Podcasts about Swe

Google just dropped Gemini 3 Flash—a model that outperforms Gemini 2.5 Pro (their last top model) while running 3x faster at less than 1/4 the cost. It's frontier-level reasoning at Flash-level speed, and it's rolling out globally right now.We're sitting down with Logan Kilpatrick from Google DeepMind to explore what this actually means for developers, knowledge workers, and anyone trying to figure out how AI fits into their workflow.What we'll cover:

Carolina Caro, WE Local Portland keynote and CEO of Conscious Leadership Partners, breaks down communication blind spots for engineers and explores how your communication style shapes your leadership impact in this episode. As a scientist by training, Carolina reflects on why technical expertise isn't enough and speaks to the platinum rule, where you meet people where they are and communicate in the way they can best receive. In conversation with FY26 SWE President Inaas Darrat, hear how to strengthen your communication without compromising your authenticity and why communication style is an often-overlooked dimension of diversity. Carolina will expand on these insights as the opening keynote at WE Local Portland, taking place Feb. 27-28. WE Local conferences bring together engineers and technologists for networking opportunities, professional development, and inspirational speakers. Register at welocal.swe.org to join SWE at an upcoming WE Local near you! — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

CoreStory is building code intelligence platforms that address the fundamental limitation of today's coding agents: their inability to navigate complex enterprise codebases. While foundation models excel at greenfield development, they fail at real-world engineering tasks in systems spanning millions of lines of code. CoreStory's context layer delivers a 44% improvement on SWE-bench, the industry's standard benchmark for measuring coding agent effectiveness on actual GitHub issues. In this episode of BUILDERS, I sat down with Anand Kulkarni, CEO of CoreStory, to explore how his team is enabling the shift to AI-native engineering and seeding the category of spec-driven development across Microsoft, GitHub, and Amazon. Topics Discussed: Building with GPT-3 API 18 months before ChatGPT went public Why even GPT-5 and Opus 4.5 struggle with enterprise codebases on SWE-bench The narrative shift required when selling AI pre- and post-ChatGPT CoreStory's 44% improvement in coding agent performance through context intelligence How "spec-driven development" got adopted by Microsoft, GitHub, and Amazon without formal analyst relations The parallel between JIRA monetizing Agile and CoreStory enabling AI-native engineering Three-channel distribution: direct enterprise, coding agent partnerships via MCP, and hyperscaler/GSI routes Why specs become the source of truth while code becomes disposable in the AI era GTM Lessons For B2B Founders: Match your narrative precision to technical depth: CoreStory deploys three distinct positioning strategies based on audience sophistication. For AI practitioners tracking benchmarks, they lead with "44% SWE-bench improvement"—a metric that immediately signals meaningful progress on the hardest problem in the space. For engineering leaders aware of AI tooling but not deep in the research, they focus on velocity gains and ROI metrics. For executives, they describe reverse-engineering codebases into machine-readable specs. The key insight: technical audiences dismiss vague value props, while non-technical audiences get lost in benchmark details. Map your positioning to how your audience measures success in their world. Seed category language through earned adoption, not manufactured consensus: Anand initially called their approach "requirements-driven development" before simplifying to "spec-driven development." Rather than pitching analysts, they used the term consistently in customer conversations, gave talks at GitHub Universe, and shipped demos showing the workflow. When customers naturally adopted the language and community leaders began using similar terminology independently, Microsoft and GitHub followed with their own implementations (like GitHub's SpecKit). The lesson: category language sticks when practitioners choose to use it because it clarifies their work, not because a vendor pushed it. Focus on customer adoption as proof of concept before seeking broader market validation. Position against emergent practices, not just incumbent products: CoreStory doesn't position against legacy code analysis tools—they position as the enabler of AI-native engineering, the discipline that will displace Agile. Anand's insight from watching JIRA's success: "People don't love JIRA. What they love is Agile as a way to move away from waterfall." CoreStory is betting that 10x velocity gains from AI-native practices will drive the same categorical shift. When you're early in a technology wave, attach to the practice change (how teams will work differently) rather than feature comparisons with existing tools. Movements create markets. Design channel strategy around customer problem awareness: CoreStory's three channels map to different stages of buyer sophistication. Direct enterprise comes from teams already deep in AI engineering who've hit the context limitation wall. Coding agent partnerships (via MCP integration with tools like Cognition and Factory) serve builders wanting better AI tooling who haven't diagnosed the context problem yet. Hyperscalers and GSIs distribute into modernization and maintenance projects where AI enablement is emerging as a requirement. Each channel serves a distinct buyer journey stage. Don't force one go-to-market motion—design multiple paths based on where different customer segments are in understanding the problem you solve. Navigate pre-legitimacy markets by hiding the breakthrough: Before ChatGPT, selling anything AI-driven faced immediate skepticism about whether it was "real" or just smoke and mirrors. Anand couldn't lead with AI without triggering disbelief. CoreStory focused on delivered outcomes—"here's what you'll be able to do"—with AI as the mechanism, not the message. Post-ChatGPT, the challenge flipped: everyone expects AI, but now the differentiation question becomes harder. If you're building on emerging technology before market consensus forms, deemphasize the technology until buyers have context to evaluate it. Once the market validates the technology category, shift to demonstrating your specific technical advantage within it. // Sponsors: Front Lines — We help B2B tech companies launch, manage, and grow podcasts that drive demand, awareness, and thought leadership. www.FrontLines.io The Global Talent Co. — We help tech startups find, vet, hire, pay, and retain amazing marketing talent that costs 50-70% less than the US & Europe. www.GlobalTalent.co // Don't Miss: New Podcast Series — How I Hire Senior GTM leaders share the tactical hiring frameworks they use to build winning revenue teams. Hosted by Andy Mowat, who scaled 4 unicorns from $10M to $100M+ ARR and launched Whispered to help executives find their next role. Subscribe here: https://open.spotify.com/show/53yCHlPfLSMFimtv0riPyM

Hey, it's Alex, let me catch you up! Since last week, OpenAI convinced OpenClaw founder Peter Steinberger to join them, while keeping OpenClaw.. well... open. Anthropic dropped Sonnet 4.6 which nearly outperforms the previous Opus and is much cheaper, Qwen released 3.5 on Chinese New Year's Eve, while DeepSeek was silent and Elon and XAI folks deployed Grok 4.20 without any benchmarks, and it's 4 500B models in a trenchcoat? Also, Anthropic updated rules state that it's breaking ToS to use their plans for anything except Claude Code & Claude SDK (and then clarified that it's OK? we're not sure) Then Google decided to drop their Gemini 3.1 Pro preview right at the start of our show, and it's very nearly the best LLM folks can use right now (though it didn't pass Nisten's vibe checks) Also, Google released Lyria 3 for music gen (though only 30 seconds?) and our own Ryan Carson blew up on X again with over 1M views for his Code Factory article, Wolfram did a deep dive into Terminal Bench and .. we have a brand new website: https://thursdai.news

Julie Daugherty, engineering associate and process project leader at Corning Incorporated, joins us live at WE25 to discuss how introverts can embrace their strengths and turn them into career superpowers. In conversation with Larry Guthrie, director of content strategy at SWE, hear how planning ahead builds confidence, how to survive “mandatory fun” networking events, and why finding extroverted champions can be key to career growth in STEM. You'll also learn how managers can support introverted engineers, what it means to be an ambivert, and why personality diversity leads to stronger teams and better problem-solving. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

I detta avsnitt av Kulturbarnen berättar Pontus om projektet han drivit i smyg i snart ett år, re/works SWE för Epidemic Sound, i vilket han bjudit in ett stort antal svenska artister och låtit dem omtolka musik från ett stort låtbibliotek dedikerat för synk och ljudläggning av internet. Välkomna!

Hey dear subscriber, Alex here from W&B, let me catch you up! This week started with Anthropic releasing /fast mode for Opus 4.6, continued with ByteDance reality-shattering video model called SeeDance 2.0, and then the open weights folks pulled up! Z.ai releasing GLM-5, a 744B top ranking coder beast, and then today MiniMax dropping a heavily RL'd MiniMax M2.5, showing 80.2% on SWE-bench, nearly beating Opus 4.6! I've interviewed Lou from Z.AI and Olive from MiniMax on the show today back to back btw, very interesting conversations, starting after TL;DR!So while the OpenSource models were catching up to frontier, OpenAI and Google both dropped breaking news (again, during the show), with Gemini 3 Deep Think shattering the ArcAGI 2 (84.6%) and Humanity's Last Exam (48% w/o tools)... Just an absolute beast of a model update, and OpenAI launched their Cerebras collaboration, with GPT 5.3 Codex Spark, supposedly running at over 1000 tokens per second (but not as smart) Also, crazy week for us at W&B as we scrambled to host GLM-5 at day of release, and are working on dropping Kimi K2.5 and MiniMax both on our inference service! As always, all show notes in the end, let's DIVE IN! ThursdAI - AI is speeding up, don't get left behind! Sub and I'll keep you up to date with a weekly catch upOpen Source LLMsZ.ai launches GLM-5 - #1 open-weights coder with 744B parameters (X, HF, W&B inference)The breakaway open-source model of the week is undeniably GLM-5 from Z.ai (formerly known to many of us as Zhipu AI). We were honored to have Lou, the Head of DevRel at Z.ai, join us live on the show at 1:00 AM Shanghai time to break down this monster of a release.GLM-5 is massive, not something you run at home (hey, that's what W&B inference is for!) but it's absolutely a model that's worth thinking about if your company has on prem requirements and can't share code with OpenAI or Anthropic. They jumped from 355B in GLM4.5 and expanded their pre-training data to a whopping 28.5T tokens to get these results. But Lou explained that it's not only about data, they adopted DeepSeeks sparse attention (DSA) to help preserve deep reasoning over long contexts (this one has 200K)Lou summed up the generational leap from version 4.5 to 5 perfectly in four words: “Bigger, faster, better, and cheaper.” I dunno about faster, this may be one of those models that you hand off more difficult tasks to, but definitely cheaper, with $1 input/$3.20 output per 1M tokens on W&B! While the evaluations are ongoing, the one interesting tid-bit from Artificial Analysis was, this model scores the lowest on their hallucination rate bench! Think about this for a second, this model is neck-in-neck with Opus 4.5, and if Anthropic didn't release Opus 4.6 just last week, this would be an open weights model that rivals Opus! One of the best models the western foundational labs with all their investments has out there. Absolutely insane times. MiniMax drops M2.5 - 80.2% on SWE-bench verified with just 10B active parameters (X, Blog)Just as we wrapped up our conversation with Lou, MiniMax dropped their release (though not weights yet, we're waiting ⏰) and then Olive Song, a senior RL researcher on the team, joined the pod, and she was an absolute wealth of knowledge! Olive shared that they achieved an unbelievable 80.2% on SWE-Bench Verified. Digest this for a second: a 10B active parameter open-source model is directly trading blows with Claude Opus 4.6 (80.8%) on the one of the hardest real-world software engineering benchmark we currently have. While being alex checks notes ... 20X cheaper and much faster to run? Apparently their fast version gets up to 100 tokens/s. Olive shared the “not so secret” sauce behind this punch-above-its-weight performance. The massive leap in intelligence comes entirely from their highly decoupled Reinforcement Learning framework called “Forge.” They heavily optimized not just for correct answers, but for the end-to-end time of task performing. In the era of bloated reasoning models that spit out ten thousand “thinking” tokens before writing a line of code, MiniMax trained their model across thousands of diverse environments to use fewer tools, think more efficiently, and execute plans faster. As Olive noted, less time waiting and fewer tools called means less money spent by the user. (as confirmed by @swyx at the Windsurf leaderboard, developers often prefer fast but good enough models) I really enjoyed the interview with Olive, really recommend you listen to the whole conversation starting at 00:26:15. Kudos MiniMax on the release (and I'll keep you updated when we add this model to our inference service) Big Labs and breaking newsThere's a reason the show is called ThursdAI, and today this reason is more clear than ever, AI biggest updates happen on a Thursday, often live during the show. This happened 2 times last week and 3 times today, first with MiniMax and then with both Google and OpenAI! Google previews Gemini 3 Deep Think, top reasoning intelligence SOTA Arc AGI 2 at 84% & SOTA HLE 48.4% (X , Blog)I literally went

Hey, Alex from W&B here

Zoe, SWENext Influencer and FIRST Robotics leader, and Charlee, a member of the 2025 STEM Next Flight Crew youth ambassador program, join us live from the WE25 Diverse Podcast Studio in New Orleans to share their experiences as young STEM leaders. In conversation with Larry Guthrie, director of content strategy at SWE, they reflect on how Invent It. Build It. sparked powerful connections, what leadership looks like as a precollege student, and how they navigate bias, responsibility, and pressure as “the first” in their communities. Hear their experiences starting STEM clubs and expanding access in rural areas, plus why a strong community is critical to shaping a more inclusive future in engineering. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

This is a link post. Improving model performance by scaling up inference compute is the next big thing in frontier AI. But the charts being used to trumpet this new paradigm can be misleading. While they initially appear to show steady scaling and impressive performance for models like o1 and o3, they really show poor scaling (characteristic of brute force) and little evidence of improvement between o1 and o3. I explore how to interpret these new charts and what evidence for strong scaling and progress would look like. From scaling training to scaling inference The dominant trend in frontier AI over the last few years has been the rapid scale-up of training — using more and more compute to produce smarter and smarter models. Since GPT-4, this kind of scaling has run into challenges, so we haven't yet seen models much larger than GPT-4. But we have seen a recent shift towards scaling up the compute used during deployment (aka 'test-time compute' or ‘inference compute'), with more inference compute producing smarter models. You could think of this as a change in strategy from improving the quality of your employees' work via giving them more years of training in which acquire [...] --- First published: February 2nd, 2026 Source: https://forum.effectivealtruism.org/posts/zNymXezwySidkeRun/inference-scaling-and-the-log-x-chart Linkpost URL:https://www.tobyord.com/writing/inference-scaling-and-the-log-x-chart --- Narrated by TYPE III AUDIO. ---Images from the article:

Hey guys, Alex here

This episode is sponsored by BD. Showing up as your authentic self in engineering isn't always easy. In this conversation, Christine Kearney Hawkins, senior staff R&D engineer in BD's Peripheral Intervention business and SWE life member, shares her 20+ year journey navigating authenticity, leadership, and innovation in STEM. From being told that she couldn't be both an engineer and a mom, to learning that her bubbly enthusiasm is a strength and not a liability, Christine reflects on how embracing who she is shaped her career and impact. In conversation with host Sam East, hear how authentic leadership fuels better innovation outcomes, what to do when workplace feedback conflicts with your core values, and practical advice to create cultures where people feel safe bringing their whole selves to work. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Hey! Alex here, with another weekly AI update! It seems like ThursdAI is taking a new direction, as this is our 3rd show this year, and a 3rd deep dive into topics (previously Ralph, Agent Skills), please let me know if the comments if you like this format. This week's deep dive is into Clawdbot, a personal AI assistant you install on your computer, but can control through your phone, has access to your files, is able to write code, help organize your life, but most importantly, it can self improve. Seeing Wolfred (my Clawdbot) learn to transcribe incoming voice messages blew my mind, and I wanted to share this one with you at length! We had Dan Peguine on the show for the deep dive + both Wolfram and Yam are avid users! This one is not to be missed. If ThursdAI is usually too technical for you, use Claude, and install Clawdbot after you read/listen to the deep dive!Also this week, we read Claude's Constitution that Anthropic released, heard a bunch of new TTS models (some are open source and very impressive) and talked about the new lightspeed coding model GLM 4.7 Flash. First the news, then deep dive, lets go

Stepping into STEM leadership doesn't require a senior title or having everything figured out. In this episode, Katie Ashley, volunteer coordinator of the SWE Early Career Professionals Affinity Group (SWE ECP AG), is joined by Zoe Husted, ECP AG conferences and awards co-chair and president of the SWE Golden Gate Section, and Kathryn Wittek, ECP AG design coordinator and president of the SWE Baltimore-Washington Section, to explore what leadership can look like in the beginning years of an engineering career. Drawing from their experiences in SWE, team sports, and technical roles, Zoe and Kathryn share when they started seeing themselves as leaders, how they navigate leading more experienced colleagues, and why learning and leading often happen at the same time. Hear their tips to find community as an early-career engineer, plus how the skills they have developed through SWE have translated into the workplace. The SWE ECP AG was formed to equip individuals with the support, resources, and inclusive community to excel in the first ten years of their career. Get involved and find out about upcoming ECP AG events at https://earlycareerprofessionalsag.swe.org/. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

 CZE looking for the first WJC gold in 25 years. SWE looking for just it's second in 45 years! We put a bow on the 2026 IIHF World Junior Championships!

*This is an all-time game in the World Junior Championships Tournament history*To-Know: SWE has lost four-straight semis. FIN has made four-straight. All-time matchup numbers are 21 wins for SWE and 20 for Finland. Enjoy!

From creating SWE-bench in a Princeton basement to shipping CodeClash, SWE-bench Multimodal, and SWE-bench Multilingual, John Yang has spent the last year and a half watching his benchmark become the de facto standard for evaluating AI coding agents—trusted by Cognition (Devin), OpenAI, Anthropic, and every major lab racing to solve software engineering at scale. We caught up with John live at NeurIPS 2025 to dig into the state of code evals heading into 2026: why SWE-bench went from ignored (October 2023) to the industry standard after Devin's launch (and how Walden emailed him two weeks before the big reveal), how the benchmark evolved from Django-heavy to nine languages across 40 repos (JavaScript, Rust, Java, C, Ruby), why unit tests as verification are limiting and long-running agent tournaments might be the future (CodeClash: agents maintain codebases, compete in arenas, and iterate over multiple rounds), the proliferation of SWE-bench variants (SWE-bench Pro, SWE-bench Live, SWE-Efficiency, AlgoTune, SciCode) and how benchmark authors are now justifying their splits with curation techniques instead of just “more repos,” why Tau-bench's “impossible tasks” controversy is actually a feature not a bug (intentionally including impossible tasks flags cheating), the tension between long autonomy (5-hour runs) vs. interactivity (Cognition's emphasis on fast back-and-forth), how Terminal-bench unlocked creativity by letting PhD students and non-coders design environments beyond GitHub issues and PRs, the academic data problem (companies like Cognition and Cursor have rich user interaction data, academics need user simulators or compelling products like LMArena to get similar signal), and his vision for CodeClash as a testbed for human-AI collaboration—freeze model capability, vary the collaboration setup (solo agent, multi-agent, human+agent), and measure how interaction patterns change as models climb the ladder from code completion to full codebase reasoning.We discuss:* John's path: Princeton → SWE-bench (October 2023) → Stanford PhD with Diyi Yang and the Iris Group, focusing on code evals, human-AI collaboration, and long-running agent benchmarks* The SWE-bench origin story: released October 2023, mostly ignored until Cognition's Devin launch kicked off the arms race (Walden emailed John two weeks before: “we have a good number”)* SWE-bench Verified: the curated, high-quality split that became the standard for serious evals* SWE-bench Multimodal and Multilingual: nine languages (JavaScript, Rust, Java, C, Ruby) across 40 repos, moving beyond the Django-heavy original distribution* The SWE-bench Pro controversy: independent authors used the “SWE-bench” name without John's blessing, but he's okay with it (”congrats to them, it's a great benchmark”)* CodeClash: John's new benchmark for long-horizon development—agents maintain their own codebases, edit and improve them each round, then compete in arenas (programming games like Halite, economic tasks like GDP optimization)* SWE-Efficiency (Jeffrey Maugh, John's high school classmate): optimize code for speed without changing behavior (parallelization, SIMD operations)* AlgoTune, SciCode, Terminal-bench, Tau-bench, SecBench, SRE-bench: the Cambrian explosion of code evals, each diving into different domains (security, SRE, science, user simulation)* The Tau-bench “impossible tasks” debate: some tasks are underspecified or impossible, but John thinks that's actually a feature (flags cheating if you score above 75%)* Cognition's research focus: codebase understanding (retrieval++), helping humans understand their own codebases, and automatic context engineering for LLMs (research sub-agents)* The vision: CodeClash as a testbed for human-AI collaboration—vary the setup (solo agent, multi-agent, human+agent), freeze model capability, and measure how interaction changes as models improve—John Yang* SWE-bench: https://www.swebench.com* X: https://x.com/jyangballinFull Video EpisodeTimestamps00:00:00 Introduction: John Yang on SWE-bench and Code Evaluations00:00:31 SWE-bench Origins and Devon's Impact on the Coding Agent Arms Race00:01:09 SWE-bench Ecosystem: Verified, Pro, Multimodal, and Multilingual Variants00:02:17 Moving Beyond Django: Diversifying Code Evaluation Repositories00:03:08 Code Clash: Long-Horizon Development Through Programming Tournaments00:04:41 From Halite to Economic Value: Designing Competitive Coding Arenas00:06:04 Ofir's Lab: SWE-ficiency, AlgoTune, and SciCode for Scientific Computing00:07:52 The Benchmark Landscape: TAU-bench, Terminal-bench, and User Simulation00:09:20 The Impossible Task Debate: Refusals, Ambiguity, and Benchmark Integrity00:12:32 The Future of Code Evals: Long Autonomy vs Human-AI Collaboration00:14:37 Call to Action: User Interaction Data and Codebase Understanding Research Get full access to Latent.Space at www.latent.space/subscribe

From creating SWE-bench in a Princeton basement to shipping CodeClash, SWE-bench Multimodal, and SWE-bench Multilingual, John Yang has spent the last year and a half watching his benchmark become the de facto standard for evaluating AI coding agents—trusted by Cognition (Devin), OpenAI, Anthropic, and every major lab racing to solve software engineering at scale. We caught up with John live at NeurIPS 2025 to dig into the state of code evals heading into 2026: why SWE-bench went from ignored (October 2023) to the industry standard after Devin's launch (and how Walden emailed him two weeks before the big reveal), how the benchmark evolved from Django-heavy to nine languages across 40 repos (JavaScript, Rust, Java, C, Ruby), why unit tests as verification are limiting and long-running agent tournaments might be the future (CodeClash: agents maintain codebases, compete in arenas, and iterate over multiple rounds), the proliferation of SWE-bench variants (SWE-bench Pro, SWE-bench Live, SWE-Efficiency, AlgoTune, SciCode) and how benchmark authors are now justifying their splits with curation techniques instead of just "more repos," why Tau-bench's "impossible tasks" controversy is actually a feature not a bug (intentionally including impossible tasks flags cheating), the tension between long autonomy (5-hour runs) vs. interactivity (Cognition's emphasis on fast back-and-forth), how Terminal-bench unlocked creativity by letting PhD students and non-coders design environments beyond GitHub issues and PRs, the academic data problem (companies like Cognition and Cursor have rich user interaction data, academics need user simulators or compelling products like LMArena to get similar signal), and his vision for CodeClash as a testbed for human-AI collaboration—freeze model capability, vary the collaboration setup (solo agent, multi-agent, human+agent), and measure how interaction patterns change as models climb the ladder from code completion to full codebase reasoning. We discuss: John's path: Princeton → SWE-bench (October 2023) → Stanford PhD with Diyi Yang and the Iris Group, focusing on code evals, human-AI collaboration, and long-running agent benchmarks The SWE-bench origin story: released October 2023, mostly ignored until Cognition's Devin launch kicked off the arms race (Walden emailed John two weeks before: "we have a good number") SWE-bench Verified: the curated, high-quality split that became the standard for serious evals SWE-bench Multimodal and Multilingual: nine languages (JavaScript, Rust, Java, C, Ruby) across 40 repos, moving beyond the Django-heavy original distribution The SWE-bench Pro controversy: independent authors used the "SWE-bench" name without John's blessing, but he's okay with it ("congrats to them, it's a great benchmark") CodeClash: John's new benchmark for long-horizon development—agents maintain their own codebases, edit and improve them each round, then compete in arenas (programming games like Halite, economic tasks like GDP optimization) SWE-Efficiency (Jeffrey Maugh, John's high school classmate): optimize code for speed without changing behavior (parallelization, SIMD operations) AlgoTune, SciCode, Terminal-bench, Tau-bench, SecBench, SRE-bench: the Cambrian explosion of code evals, each diving into different domains (security, SRE, science, user simulation) The Tau-bench "impossible tasks" debate: some tasks are underspecified or impossible, but John thinks that's actually a feature (flags cheating if you score above 75%) Cognition's research focus: codebase understanding (retrieval++), helping humans understand their own codebases, and automatic context engineering for LLMs (research sub-agents) The vision: CodeClash as a testbed for human-AI collaboration—vary the setup (solo agent, multi-agent, human+agent), freeze model capability, and measure how interaction changes as models improve — John Yang SWE-bench: https://www.swebench.com X: https://x.com/jyangballin Chapters 00:00:00 Introduction: John Yang on SWE-bench and Code Evaluations 00:00:31 SWE-bench Origins and Devon's Impact on the Coding Agent Arms Race 00:01:09 SWE-bench Ecosystem: Verified, Pro, Multimodal, and Multilingual Variants 00:02:17 Moving Beyond Django: Diversifying Code Evaluation Repositories 00:03:08 Code Clash: Long-Horizon Development Through Programming Tournaments 00:04:41 From Halite to Economic Value: Designing Competitive Coding Arenas 00:06:04 Ofir's Lab: SWE-ficiency, AlgoTune, and SciCode for Scientific Computing 00:07:52 The Benchmark Landscape: TAU-bench, Terminal-bench, and User Simulation 00:09:20 The Impossible Task Debate: Refusals, Ambiguity, and Benchmark Integrity 00:12:32 The Future of Code Evals: Long Autonomy vs Human-AI Collaboration 00:14:37 Call to Action: User Interaction Data and Codebase Understanding Research

Karen Horting, Executive Director and CEO of the Society of Women Engineers, talks about SWE's archives at the Reuther Library and shares how the 75-year-old organization leverages its history to advocate for the inclusion of women in science, technology, engineering, and mathematics (STEM). Related Resources: Society of Women Engineers 75th Anniversary SWE Archives Virtual Tour [Part 1] SWE Archives Virtual Tour [Part 2] Related Collections: Society of Women Engineers Records (LR001539) Society of Women Engineers Publications (LR002487) Episode Credits Interviewee: Karen Horting Producers: Dan Golodner and Troy Eller English Music: Bart Bealmear

Emily breaks down her 2025 Holiday Gift Guide through the lens of the five pillars of sexual intelligence—embodiment, health, self-knowledge, self-acceptance, and collaboration. She explores how shifting from performance to presence can transform your relationship with pleasure, offering curated tools and practices that help you slow down, feel your body, and understand yourself as a sexual being. This episode is for anyone ready to move beyond quick fixes and embrace a more holistic approach to sexuality, turning intimacy into something that nourishes your whole system. In this episode, you'll learn: • The five pillars of Sex IQ create a holistic framework for understanding your sexuality and taking responsibility for your pleasure—moving you from performance to presence • Embodiment practices like breathwork and sonic wave toys help you overcome the mind-body disconnect during sex by bringing you back to physical sensations instead of staying in your head worrying about how you look or what you need to do • Self-acceptance isn't about waiting for your body to change—it's about recognizing and reframing the pleasure thieves (stress, trauma, and shame) that tell you you're unworthy, and actively replacing negative thought patterns with affirmations that honor your body right now More Dr. Emily:  • The 2025 Sex With Emily Holiday Gift Guide • Apply for Emily's 1:1 Coaching Opportunity HERE or reach out to enrollment@sexwithemily.com for more information  • Shop With Emily! Explore Emily's favorite toys, pleasure accessories, bedroom essentials, and more — designed to support your pleasure and confidence. Free shipping on orders $99+ (some exclusions apply). • Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. • Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides. • The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure • Want more? Visit the Sex With Emily Website • Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube • Let's text: Sign up here • Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. Shop the Holiday Gift Guide now! (See the full Gift Guide HERE)  Embodiment: • LELO SONA 3- Use code EMILY20 for 20% on top of ongoing sales at lelo.com • Common Confidential Massage Butter- Use code SEXWITHEMILY for 15% off at Commonconfidential.com. • Cornbread Hemp - Use code SWE for a discount at cornbreadhemp.com/swe  Health: • Bathmate Hydro Pump-  Use SWE10 for 10% off your order at bathmatedirect.com • HigherDOSE Infrared Sauna Blanket- Head to higherdose.com • V-Health Vaginal Rejuvenation Gel- Use code EMILY10 for 10% off at getvhealth.com  • Kroma Wellness Beauty Matcha Latte - Use code SEXWITHEMILY for 15% off at kromawellness.com  Collaboration:  • Promescent Delay Spray- Head to promescent.com  • LELO F2S- Use code EMILY20 for 20% on top of ongoing sales at lelo.com  • Crave Leather Handcuffs- Head to shop.sexwithemily.com/crave  Self- Knowledge: • Magic Wand Mini - Head to shop.sexwithemily.com/magicwand  • Je Joue Hera Flex Rabbit Vibrator - Go to sexwithemily.com/hera and use code EMILY20 for 20% off  • SmartSX- Head to sexwithemily.com/smartsx  Self- Acceptance:  • The Class - Head to theclass.com  • Droplette- Head to droplette.io This episode is sponsored by…  Biolouve- Get 15% off with code EMILYPOWERMOVE @ https://biolouve.com/  Timestamps 0:00 - Introduction 1:25 - The 5 Pillars of Sex IQ Framework 2:36 - Pillar 1: Embodiment 10:23 - Pillar 2: Health 17:52 - Pillar 3: Collaboration 24:44 - Pillar 4: Self-Knowledge 29:39 - Pillar 5: Self-Acceptance 34:21 - Wrap-Up

This episode is sponsored by Bechtel. Josefina Alvarez and Kira McKay, supplier quality representatives at Bechtel, sit down with SWE President-Elect Kerrie Greenfelder to discuss how they landed their first jobs through the SWE and SHPE career fairs and also discovered a side of engineering they never knew existed. Recorded live at the WE25 Diverse Podcast Studio in New Orleans, hear about the behind-the-scenes components of engineering — from supply chain to quality systems — and how these roles make iconic projects possible. Kira and Josefina share candid advice for engineering students and new grads, what they've learned inside the Bechtel Supplier Quality and Expediting (BSQE) program, and how mentorship, curiosity, and saying “yes” to unfamiliar paths shaped their early careers. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

This episode features Dianne Na Penn, a senior product leader at Anthropic, discussing the launch of Claude Opus 4.5 and the evolution of frontier AI models. The conversation explores how Anthropic approaches model development—balancing ambitious capability roadmaps with user feedback, making strategic bets on areas like agentic coding and computer use while deliberately avoiding others like image generation. Dianne shares insights on the shifting nature of AI evaluation (moving beyond saturated benchmarks like SWE-bench toward more open-ended measures), the evolution of scaffolding from "training wheels" to intelligence amplifiers, and why she believes we're closer to transformative long-running AI than most people think. She also discusses Anthropic's distinctive culture of authenticity, the under appreciated benefits of model alignment for producing independent-thinking AI, and why the real bottleneck to AI agents isn't model capability anymore but product innovation. (0:00) Intro(0:57) Starting the Work on Opus 4.5(2:04) Model Capabilities and Surprises(5:59) Computer Use and Practical Applications(7:21) Pricing and Positioning(10:02) Customer Feedback and Early Access(16:44) The Reality of Enterprise Agents(18:47) Future of AI and Long-Running Intelligence(28:06) Anthropic's Culture and Decision Making(30:31) Key Decisions and Fun Moments(33:45) Quickfire With your co-hosts: @jacobeffron - Partner at Redpoint, Former PM Flatiron Health @patrickachase - Partner at Redpoint, Former ML Engineer LinkedIn @ericabrescia - Former COO Github, Founder Bitnami (acq'd by VMWare) @jordan_segall - Partner at Redpoint

The common advice for career growth is to “move up” into management — but what if your true passion lies in staying close to the technology itself? In this episode, host Sam East speaks with Deb Whitis, Ph.D., and Amrita Maguire, both of the SWE Technical Career Path Affinity Group, about what it means to grow, lead, and make an impact as engineers while staying on the technical career path. From developing nickel-based superalloys that power jet engines to advancing ergonomic standards and AI-enabled design, Deb and Amrita reflect on their careers and share how technical leadership can be just as influential as managing people. They also highlight the work of SWE's Technical Career Path Affinity Group, including a new mentorship program helping women chart their own path as innovators, inventors, and subject matter experts. Get involved with the SWE Technical Career Path Affinity Group: https://affinitygroups.swe.org/technical-career-path/ — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Google just released Gemini 3, and I tested it with 3 real business use cases: prepping for a $200K sales call, building an interactive sales dashboard, and creating a complete website from scratch—all in minutes.In this video, I show you LIVE examples of:✅ Gemini Agent prepping an entire sales call (research, pitch angles, objection handling, follow-ups)✅ Dynamic dashboards with real-time calculations and interactive sliders✅ Full website generation with 921 lines of working code in under 60 seconds✅ Custom image generation and prompt engineeringThis isn't theory—I'm screen recording everything as it happens so you can see the actual speed and quality.⏱️ TIMESTAMPS:0:00 - Intro: Why Gemini 3 is a Big Deal1:15 - Benchmark Breakdown (Why This Matters)2:27 - Use Case #1: $200K Sales Call Prep4:46 - Visual Layout & Interactive Infographic Demo7:03 - Use Case #2: Building a Website from Scratch (Live Code)9:27 - Use Case #3: Interactive Q4 Sales Dashboard11:51 - Testing the Prompt Generator Live12:45 - Final Thoughts & Why This Changes Everything

This week on the show Maine recaps the 2025 ADCC North American trials live from Orlando Florida.  We recap all the divisions starting with the Men'sWe discuss the -66.0 kg Dorian Olivarez-77.0 kg Jacob Bornemann-88.0 kg Jon Blank-99.0 kg Achilles Rocha+99.0 kg Brandon ReedThen  move into the recap of the womens division-55.0 kgAna Mayordomo-65.0 kg Morgan "Mo" Black+65.0 kg Maia MatalonRecorded 11-17-2025

Dans cet épisode, Emmanuel, Katia et Guillaume discutent de Spring 7, Quarkus, d'Infinispan et Keycloak. On discute aussi de projets sympas comme Javelit, de comment démarre une JVM, du besoin d'argent de NTP. Et puis on discute du changement de carrière d'Emmanuel. Enregistré le 14 novembre 2025 Téléchargement de l'épisode LesCastCodeurs-Episode-332.mp3 ou en vidéo sur YouTube. News Emmanuel quitte Red Hat après 20 ans https://emmanuelbernard.com/blog/2025/11/13/leaving-redhat/ Langages Support HTTP/3 dans le HttpClient de JDK 26 - https://inside.java/2025/10/22/http3-support/ JDK 26 introduit le support de HTTP/3 dans l'API HttpClient existante depuis Java 11 HTTP/3 utilise le protocole QUIC sur UDP au lieu de TCP utilisé par HTTP/2 Par défaut HttpClient préfère HTTP/2, il faut explicitement configurer HTTP/3 avec Version.HTTP_3 Le client effectue automatiquement un downgrade vers HTTP/2 puis HTTP/1.1 si le serveur ne supporte pas HTTP/3 On peut forcer l'utilisation exclusive de HTTP/3 avec l'option H3_DISCOVERY en mode HTTP_3_URI_ONLY HttpClient apprend qu'un serveur supporte HTTP/3 via le header alt-svc (RFC 7838) et utilise cette info pour les requêtes suivantes La première requête peut utiliser HTTP/2 même avec HTTP/3 préféré, mais la seconde utilisera HTTP/3 si le serveur l'annonce L'équipe OpenJDK encourage les tests et retours d'expérience sur les builds early access de JDK 26 Librairies Eclispe Jetty et CometD changent leurs stratégie de support https://webtide.com/end-of-life-changes-to-eclipse-jetty-and-cometd/ À partir du 1er janvier 2026, Webtide ne publiera plus Jetty 9/10/11 et CometD 5/6/7 sur Maven Central Pendant 20 ans, Webtide a financé les projets Jetty et CometD via services et support, publiant gratuitement les mises à jour EOL Le comportement des entreprises a changé : beaucoup cherchent juste du gratuit plutôt que du véritable support Des sociétés utilisent des versions de plus de 10 ans sans migrer tant que les correctifs CVE sont gratuits Cette politique gratuite a involontairement encouragé la complaisance et retardé les migrations vers versions récentes MITRE développe des changements au système CVE pour mieux gérer les concepts d'EOL Webtide lance un programme de partenariat avec TuxCare et HeroDevs pour distribuer les résolutions CVE des versions EOL Les binaires EOL seront désormais distribués uniquement aux clients commerciaux et via le réseau de partenaires Webtide continue le support standard open-source : quand Jetty 13 sortira, Jetty 12.1 recevra des mises à jour pendant 6 mois à un an Ce changement vise à clarifier la politique EOL avec une terminologie industrielle établie Améliorations cloud du SDK A2A Java https://quarkus.io/blog/quarkus-a2a-cloud-enhancements/ Version 0.3.0.Final du SDK A2A Java apporte des améliorations pour les environnements cloud et distribués Composants en mémoire remplacés par des implémentations persistantes et répliquées pour environnements multi-instances JpaDatabaseTaskStore et JpaDatabasePushNotificationConfigStore permettent la persistance des tâches et configurations en base PostgreSQL ReplicatedQueueManager assure la réplication des événements entre instances A2A Agent via Kafka et MicroProfile Reactive Messaging Exemple complet de déploiement Kubernetes avec Kind incluant PostgreSQL, Kafka via Strimzi, et load balancing entre pods Démonstration pratique montrant que les messages peuvent être traités par différents pods tout en maintenant la cohérence des tâches Architecture inspirée du SDK Python A2A, permettant la gestion de tâches asynchrones longues durée en environnement distribué Quarkus 3.29 sort avec des backends de cache multiples et support du débogueur Qute https://quarkus.io/blog/quarkus-3-29-released/ Possibilité d'utiliser plusieurs backends de cache simultanément dans une même application Chaque cache peut être associé à un backend spécifique (par exemple Caffeine et Redis ou Infinispan) Support du Debug Adapter Protocol (DAP) pour déboguer les templates Qute directement dans l'IDE et dans la version 3.28 Configuration programmatique de la protection CSRF via une API fluent Possibilité de restreindre les filtres OIDC à des flux d'authentification spécifiques avec annotations Support des dashboards Grafana personnalisés via fichiers JSON dans META-INF/grafana/ Extension Liquibase MongoDB supporte désormais plusieurs clients simultanés Amélioration significative des performances de build avec réduction des allocations mémoire Parallélisation de tâches comme la génération de proxies Hibernate ORM et la construction des Jar Et l'utilisation des fichiers .proto est plus simple dans Quarkus avbec Quarkus gRPC Zero https://quarkus.io/blog/grpc-zero/ c'est toujours galere des fichiers .proto car les generateurs demandent des executables natifs maintenant ils sont bundlés dans la JVM et vous n'avez rien a configurer cela utilise Caffeine pour faire tourner cela en WASM dans la JVM Spring AI 1.1 est presque là https://spring.io/blog/2025/11/08/spring-ai-1-1-0-RC1-available-now support des MCP tool caching pour les callback qui reduit les iooerations redondantes Access au contenu de raisonnement OpenAI Un modele de Chat MongoDB Support du modele de penser Ollama Reessaye sur les echec de reseau OpenAI speech to text Spring gRPC Les prochaines étapes pour la 1.0.0 https://spring.io/blog/2025/11/05/spring-grpc-next-steps Spring gRPC 1.0 arrive prochainement avec support de Spring Boot 4 L'intégration dans Spring Boot 4.0 est reportée, prévue pour Spring Boot 4.1 Les coordonnées Maven restent sous org.springframework.grpc pour la version 1.0 Le jar spring-grpc-test est renommé en spring-grpc-test-spring-boot-autoconfigure Les packages d'autoconfiguration changent de nom nécessitant de modifier les imports Les dépendances d'autoconfiguration seront immédiatement dépréciées après la release 1.0 Migration minimale attendue pour les projets utilisant déjà la version 0.x La version 1.0.0-RC1 sera publiée dès que possible avant la version finale Spring arrete le support reactif d'Apache Pulsar https://spring.io/blog/2025/10/29/spring-pulsar-reactive-discontinued logique d'évaluer le temps passé vs le nombre d'utilisateurs c'est cependant une tendance qu'on a vu s'accélerer Spring 7 est sorti https://spring.io/blog/2025/11/13/spring-framework-7-0-general-availability Infrastructure Infinispan 16.0 https://infinispan.org/blog/2025/11/10/infinispan-16-0 Ajout majeur : migration en ligne sans interruption pour les nœuds d'un cluster (rolling upgrades) (infinispan.org) Messages de clustering refaits avec Protocol Buffers + ProtoStream : meilleure compatibilité, schéma évolutif garanti (infinispan.org) Console Web améliorée API dédiée de gestion des schémas (SchemasAdmin) pour gérer les schémas ProtoStream à distance (infinispan.org) Module de requête (query) optimisé : support complet des agrégations (sum, avg …) dans les requêtes indexées en cluster grâce à l'intégration de Hibernate Search 8.1 (infinispan.org) Serveur : image conteneur minimalisée pour réduire la surface d'attaque (infinispan.org) démarrage plus rapide grâce à séparation du démarrage cache/serveur (infinispan.org) caches pour connecteurs (Memcached, RESP) créés à la demande (on-demand) et non à l'initiaton automatique (infinispan.org) moteur Lua 5.1 mis à jour avec corrections de vulnérabilités et opérations dangereuses désactivées (infinispan.org) Support JDK : version minimale toujours JDK 17 (infinispan.org) prise en charge des threads virtuels (virtual threads) et des fonctionnalités AOT (Ahead-of-Time) de JDK plus récentes (infinispan.org) Web Javelit, une nouvelle librairie Java inspirée de Streamlit pour faire facilement et rapidement des petites interfaces web https://glaforge.dev/posts/2025/10/24/javelit-to-create-quick-interactive-app-frontends-in-java/ Site web du projet : https://javelit.io/ Javelit : outil pour créer rapidement des applications de données (mais pas que) en Java. Simplifie le développement : élimine les tracas du frontend et de la gestion des événements. Transforme une classe Java en application web en quelques minutes. Inspiré par la simplicité de Streamlit de l'écosystème Python (ou Gradio et Mesop), mais pour Java. Développement axé sur la logique : pas de code standard répétitif (boilerplate), rechargement à chaud. Interactions faciles : les widgets retournent directement leur valeur, sans besoin de HTML/CSS/JS ou gestion d'événements. Déploiement flexible : applications autonomes ou intégrables dans des frameworks Java (Spring, Quarkus, etc.). L'article de Guillaume montre comment créer une petite interface pour créer et modifier des images avec le modèle génératif Nano Banana Un deuxième article montre comment utiliser Javelit pour créer une interface de chat avec LangChain4j https://glaforge.dev/posts/2025/10/25/creating-a-javelit-chat-interface-for-langchain4j/ Améliorer l'accessibilité avec les applis JetPack Compose https://blog.ippon.fr/2025/10/29/rendre-son-application-accessible-avec-jetpack-compose/ TalkBack est le lecteur d'écran Android qui vocalise les éléments sélectionnés pour les personnes malvoyantes Accessibility Scanner et les outils Android Studio détectent automatiquement les problèmes d'accessibilité statiques Les images fonctionnelles doivent avoir un contentDescription, les images décoratives contentDescription null Le contraste minimum requis est de 4.5:1 pour le texte normal et 3:1 pour le texte large ou les icônes Les zones cliquables doivent mesurer au minimum 48dp x 48dp pour faciliter l'interaction Les formulaires nécessitent des labels visibles permanents et non de simples placeholders qui disparaissent Modifier.semantics permet de définir l'arbre sémantique lu par les lecteurs d'écran Les propriétés mergeDescendants et traversalIndex contrôlent l'ordre et le regroupement de la lecture Diriger le navigateur Chrome avec le modèle Gemini Computer Use https://glaforge.dev/posts/2025/11/03/driving-a-web-browser-with-gemini-computer-use-model-in-java/ Objectif : Automatiser la navigation web en Java avec le modèle "Computer Use" de Gemini 2.5 Pro. Modèle "Computer Use" : Gemini analyse des captures d'écran et génère des actions d'interface (clic, saisie, etc.). Outils : Gemini API, Java, Playwright (pour l'interaction navigateur). Fonctionnement : Boucle agent où Gemini reçoit une capture, propose une action, Playwright l'exécute, puis une nouvelle capture est envoyée à Gemini. Implémentation clé : Toujours envoyer une capture d'écran à Gemini après chaque action pour qu'il comprenne l'état actuel. Défis : Lenteur, gestion des CAPTCHA et pop-ups (gérables). Potentiel : Automatisation des tâches web répétitives, création d'agents autonomes. Data et Intelligence Artificielle Apicurio ajoute le support de nouveaux schema sans reconstruire Apicurio https://www.apicur.io/blog/2025/10/27/custom-artifact-types Apicurio Registry 3.1.0 permet d'ajouter des types d'artefacts personnalisés au moment du déploiement sans recompiler le projet Supporte nativement OpenAPI, AsyncAPI, Avro, JSON Schema, Protobuf, GraphQL, WSDL et XSD Trois approches d'implémentation disponibles : classes Java pour la performance maximale, JavaScript/TypeScript pour la facilité de développement, ou webhooks pour une flexibilité totale Configuration via un simple fichier JSON pointant vers les implémentations des composants personnalisés Les scripts JavaScript sont exécutés via QuickJS dans un environnement sandboxé sécurisé Un package npm TypeScript fournit l'autocomplétion et la sécurité de type pour le développement Six composants optionnels configurables : détection automatique de type, validation, vérification de compatibilité, canonicalisation, déréférencement et recherche de références Cas d'usage typiques : formats propriétaires internes, support RAML, formats legacy comme WADL, schémas spécifiques à un domaine métier Déploiement simple via Docker en montant les fichiers de configuration et scripts comme volumes Les performances varient selon l'approche : Java offre les meilleures performances, JavaScript un bon équilibre, webhooks la flexibilité maximale Le truc interessant c'est que c'est Quarkus based et donc demandait le rebuilt donc pour eviter cela, ils ont ajouter QuickJS via Chicorey un moteur WebAssembly GPT 5.1 pour les développeurs est sorti. https://openai.com/index/gpt-5-1-for-developers/ C'est le meilleur puisque c'est le dernier :slightly_smiling_face: Raisonnement Adaptatif et Efficace : GPT-5.1 ajuste dynamiquement son temps de réflexion en fonction de la complexité de la tâche, le rendant nettement plus rapide et plus économique en jetons pour les tâches simples, tout en maintenant des performances de pointe sur les tâches difficiles. Nouveau Mode « Sans Raisonnement » : Un mode (reasoning_effort='none') a été introduit pour les cas d'utilisation sensibles à la latence, permettant une réponse plus rapide avec une intelligence élevée et une meilleure exécution des outils. Cache de Prompt Étendu : La mise en cache des invites est étendue jusqu'à 24 heures (contre quelques minutes auparavant), ce qui réduit la latence et le coût pour les interactions de longue durée (chats multi-tours, sessions de codage). Les jetons mis en cache sont 90 % moins chers. Améliorations en Codage : Le modèle offre une meilleure personnalité de codage, une qualité de code améliorée et de meilleures performances sur les tâches d'agenticité de code, atteignant 76,3 % sur SWE-bench Verified. Nouveaux Outils pour les Développeurs : Deux nouveaux outils sont introduits ( https://cookbook.openai.com/examples/build_a_coding_agent_with_gpt-5.1 ) : L'outil apply_patch pour des modifications de code plus fiables via des diffs structurés. L'outil shell qui permet au modèle de proposer et d'exécuter des commandes shell sur une machine locale, facilitant les boucles d'inspection et d'exécution. Disponibilité : GPT-5.1 (ainsi que les modèles gpt-5.1-codex) est disponible pour les développeurs sur toutes les plateformes API payantes, avec les mêmes tarifs et limites de débit que GPT-5. Comparaison de similarité d'articles et de documents avec les embedding models https://glaforge.dev/posts/2025/11/12/finding-related-articles-with-vector-embedding-models/ Principe : Convertir les articles en vecteurs numériques ; la similarité sémantique est mesurée par la proximité de ces vecteurs. Démarche : Résumé des articles via Gemini-2.5-flash. Conversion des résumés en vecteurs (embeddings) par Gemini-embedding-001. Calcul de la similarité entre vecteurs par similarité cosinus. Affichage des 3 articles les plus pertinents (>0.75) dans le frontmatter Hugo. Bilan : Approche "résumé et embedding" efficace, pragmatique et améliorant l'engagement des lecteurs. Outillage Composer : Nouveau modèle d'agent rapide pour l'ingénierie logicielle - https://cursor.com/blog/composer Composer est un modèle d'agent conçu pour l'ingénierie logicielle qui génère du code quatre fois plus rapidement que les modèles similaires Le modèle est entraîné sur de vrais défis d'ingénierie logicielle dans de grandes bases de code avec accès à des outils de recherche et d'édition Il s'agit d'un modèle de type mixture-of-experts optimisé pour des réponses interactives et rapides afin de maintenir le flux de développement L'entraînement utilise l'apprentissage par renforcement dans divers environnements de développement avec des outils comme la lecture de fichiers, l'édition, les commandes terminal et la recherche sémantique Cursor Bench est un benchmark d'évaluation basé sur de vraies demandes d'ingénieurs qui mesure la correction et le respect des abstractions du code existant Le modèle apprend automatiquement des comportements utiles comme effectuer des recherches complexes, corriger les erreurs de linter et écrire des tests unitaires L'infrastructure d'entraînement utilise PyTorch et Ray avec des kernels MXFP8 pour entraîner sur des milliers de GPUs NVIDIA Le système exécute des centaines de milliers d'environnements de codage sandboxés concurrents dans le cloud pour l'entraînement Composer est déjà utilisé quotidiennement par les développeurs de Cursor pour leur propre travail Le modèle se positionne juste derrière GPT-5 et Sonnet 4.5 en termes de performance sur les benchmarks internes Rex sur l'utilisation de l'IA pour les développeurs, un gain de productivité réel et des contextes adaptés https://mcorbin.fr/posts/2025-10-17-genai-dev/ Un développeur avec 18 ans d'expérience partage son retour sur l'IA générative après avoir changé d'avis Utilise exclusivement Claude Code dans le terminal pour coder en langage naturel Le "vibe coding" permet de générer des scripts et interfaces sans regarder le code généré Génération rapide de scripts Python pour traiter des CSV, JSON ou créer des interfaces HTML Le mode chirurgien résout des bugs complexes en one-shot, exemple avec un plugin Grafana fixé en une minute Pour le code de production, l'IA génère les couches repository, service et API de manière itérative, mais le dev controle le modele de données Le développeur relit toujours le code et ajuste manuellement ou via l'IA selon le besoin L'IA ne remplacera pas les développeurs car la réflexion, conception et expertise technique restent essentielles La construction de produits robustes, scalables et maintenables nécessite une expérience humaine L'IA libère du temps sur les tâches répétitives et permet de se concentrer sur les aspects complexes ce que je trouve interessant c'est la partie sur le code de prod effectivement, je corrige aussi beaucoup les propositions de l'IA en lui demandant de faire mieux dans tel ou tel domaine Sans guide, tout cela serait perdu Affaire a suivre un article en parallele sur le métier de designer https://blog.ippon.fr/2025/11/03/lia-ne-remplace-pas-un-designer-elle-amplifie-la-difference-entre-faire-et-bien-faire/ Plus besoin de se rappeler les racourcis dans IntelliJ idea avec l'universal entry point https://blog.jetbrains.com/idea/2025/11/universal-entry-point-a-single-entry-point-for-context-aware-coding-assistance/ IntelliJ IDEA introduit Command Completion, une nouvelle façon d'accéder aux actions de l'IDE directement depuis l'éditeur Fonctionne comme la complétion de code : tapez point (.) pour voir les actions contextuelles disponibles Tapez double point (..) pour filtrer et n'afficher que les actions disponibles Propose des corrections, refactorings, génération de code et navigation selon le contexte Complète les fonctionnalités existantes sans les remplacer : raccourcis, Alt+Enter, Search Everywhere Facilite la découverte des fonctionnalités de l'IDE sans interrompre le flux de développement En Beta dans la version 2025.2, sera activé par défaut dans 2025.3 Support actuel pour Java et Kotlin, avec actions spécifiques aux frameworks comme Spring et Hibernate Homebrew, package manage pour macOS et Linux passe en version 5 https://brew.sh/2025/11/12/homebrew-5.0.0/ Téléchargements Parallèles par Défaut : Le paramètre HOMEBREW_DOWNLOAD_CONCURRENCY=auto est activé par défaut, permettant des téléchargements concurrents pour tous les utilisateurs, avec un rapport de progression. Support Linux ARM64/AArch64 en Tier 1 : Le support pour Linux ARM64/AArch64 a été promu au niveau "Tier 1" (support officiel de premier plan). Feuille de Route pour les Dépréciations macOS : Septembre 2026 (ou plus tard) : Homebrew ne fonctionnera plus sur macOS Catalina (10.15) et versions antérieures. macOS Intel (x86_64) passera en "Tier 3" (fin du support CI et des binaires précompilés/bottles). Septembre 2027 (ou plus tard) : Homebrew ne fonctionnera plus sur macOS Big Sur (11) sur Apple Silicon ni du tout sur Intel (x86_64). Sécurité et Casks : Dépréciation des Casks sans signature de code. Désactivation des Casks échouant aux vérifications Gatekeeper en septembre 2026. Les options --no-quarantine et --quarantine sont dépréciés pour ne plus faciliter le contournement des fonctionnalités de sécurité de macOS. Nouvelles Fonctionnalités & Améliorations : Support officiel pour macOS 26 (Tahoe). brew bundle supporte désormais l'installation de packages Go via un Brewfile. Ajout de la commande brew info --sizes pour afficher la taille des formulae et casks. La commande brew search --alpine permet de chercher des packages Alpine Linux. Architecture Selon l'analyste RedMonk, Java reste très pertinent dans l'aire de l'IA et des agents https://redmonk.com/jgovernor/java-relevance-in-the-ai-era-agent-frameworks-emerge/ Java reste pertinent à l'ère de l'IA, pas besoin d'apprendre une pile technique entièrement nouvelle. Capacité d'adaptation de Java ("anticorps") aux innovations (Big Data, cloud, IA), le rendant idéal pour les contextes d'entreprise. L'écosystème JVM offre des avantages sur Python pour la logique métier et les applications sophistiquées, notamment en termes de sécurité et d'évolutivité. Embabel (par Rod Johnson, créateur de Spring) : un framework d'agents fortement typé pour JVM, visant le déterminisme des projets avant la génération de code par LLM. LangChain4J : facilite l'accès aux capacités d'IA pour les développeurs Java, s'aligne sur les modèles d'entreprise établis et permet aux LLM d'appeler des méthodes Java. Koog (Jetbrains) : framework d'agents basé sur Kotlin, typé et spécifique aux développeurs JVM/Kotlin. Akka : a pivoté pour se concentrer sur les flux de travail d'agents IA, abordant la complexité, la confiance et les coûts des agents dans les systèmes distribués. Le Model Context Protocol (MCP) est jugé insuffisant, manquant d'explicabilité, de découvrabilité, de capacité à mélanger les modèles, de garde-fous, de gestion de flux, de composabilité et d'intégration sécurisée. Les développeurs Java sont bien placés pour construire des applications compatibles IA et intégrer des agents. Des acteurs majeurs comme IBM, Red Hat et Oracle continuent d'investir massivement dans Java et son intégration avec l'IA. Sécurité AI Deepfake, Hiring … A danger réel https://www.eu-startups.com/2025/10/european-startups-get-serious-about-deepfakes-as-ai-fraud-losses-surpass-e1-3-billion/ Pertes liées aux deepfakes en Europe : > 1,3 milliard € (860 M € rien qu'en 2025). Création de deepfakes désormais possible pour quelques euros. Fraudes : faux entretiens vidéo, usurpations d'identité, arnaques diverses. Startups actives : Acoru, IdentifAI, Trustfull, Innerworks, Keyless (détection et prévention). Réglementation : AI Act et Digital Services Act imposent transparence et contrôle. Recommandations : vérifier identités, former employés, adopter authentification multi-facteurs. En lien : https://www.techmonitor.ai/technology/cybersecurity/remote-hiring-cybersecurity 1 Candidat sur 4 sera Fake en 2028 selon Gartner research https://www.gartner.com/en/newsroom/press-releases/2025-07-31-gartner-survey-shows-j[…]-percent-of-job-applicants-trust-ai-will-fairly-evaluate-them Loi, société et organisation Amazon - prévoit supprimer 30.000 postes https://www.20minutes.fr/economie/4181936-20251028-amazon-prevoit-supprimer-30-000-emplois-bureau-selon-plusieurs-medias Postes supprimés : 30 000 bureaux Part des effectifs : ~10 % des employés corporatifs Tranche confirmée : 14 000 postes Divisions touchées : RH, Opérations, Devices & Services, Cloud Motifs : sur-recrutement, bureaucratie, automatisation/IA Accompagnement : 90 jours pour poste interne + aides Non concernés : entrepôts/logistique Objectif : concentrer sur priorités stratégiques NTP a besoin d'argent https://www.ntp.org/ Il n'est que le protocole qui synchronise toutes les machines du monde La fondation https://www.nwtime.org/ recherche 11000$ pour maintenir son activité Rubrique débutant Une plongée approfondie dans le démarrage de la JVM https://inside.java/2025/01/28/jvm-start-up La JVM effectue une initialisation complexe avant d'exécuter le code : validation des arguments, détection des ressources système et sélection du garbage collector approprié Le chargement de classes suit une stratégie lazy où chaque classe charge d'abord ses dépendances dans l'ordre de déclaration, créant une chaîne d'environ 450 classes même pour un simple Hello World La liaison de classes comprend trois sous-processus : vérification de la structure, préparation avec initialisation des champs statiques à leurs valeurs par défaut, et résolution des références symboliques du Constant Pool Le CDS améliore les performances au démarrage en fournissant des classes pré-vérifiées, réduisant le travail de la JVM L'initialisation de classe exécute les initialiseurs statiques via la méthode spéciale clinit générée automatiquement par javac Le Project Leyden introduit la compilation AOT dans JDK 24 pour réduire le temps de démarrage en effectuant le chargement et la liaison de classes en avance de phase Pas si débutant finalement Conférences La liste des conférences provenant de Developers Conferences Agenda/List par Aurélie Vache et contributeurs : 12-14 novembre 2025 : Devoxx Morocco - Marrakech (Morocco) 15-16 novembre 2025 : Capitole du Libre - Toulouse (France) 19 novembre 2025 : SREday Paris 2025 Q4 - Paris (France) 19-21 novembre 2025 : Agile Grenoble - Grenoble (France) 20 novembre 2025 : OVHcloud Summit - Paris (France) 21 novembre 2025 : DevFest Paris 2025 - Paris (France) 24 novembre 2025 : Forward Data & AI Conference - Paris (France) 27 novembre 2025 : DevFest Strasbourg 2025 - Strasbourg (France) 28 novembre 2025 : DevFest Lyon - Lyon (France) 1-2 décembre 2025 : Tech Rocks Summit 2025 - Paris (France) 4-5 décembre 2025 : Agile Tour Rennes - Rennes (France) 5 décembre 2025 : DevFest Dijon 2025 - Dijon (France) 9-11 décembre 2025 : APIdays Paris - Paris (France) 9-11 décembre 2025 : Green IO Paris - Paris (France) 10-11 décembre 2025 : Devops REX - Paris (France) 10-11 décembre 2025 : Open Source Experience - Paris (France) 11 décembre 2025 : Normandie.ai 2025 - Rouen (France) 14-17 janvier 2026 : SnowCamp 2026 - Grenoble (France) 22 janvier 2026 : DevCon #26 : sécurité / post-quantique / hacking - Paris (France) 29-31 janvier 2026 : Epitech Summit 2026 - Paris - Paris (France) 2-5 février 2026 : Epitech Summit 2026 - Moulins - Moulins (France) 2-6 février 2026 : Web Days Convention - Aix-en-Provence (France) 3 février 2026 : Cloud Native Days France 2026 - Paris (France) 3-4 février 2026 : Epitech Summit 2026 - Lille - Lille (France) 3-4 février 2026 : Epitech Summit 2026 - Mulhouse - Mulhouse (France) 3-4 février 2026 : Epitech Summit 2026 - Nancy - Nancy (France) 3-4 février 2026 : Epitech Summit 2026 - Nantes - Nantes (France) 3-4 février 2026 : Epitech Summit 2026 - Marseille - Marseille (France) 3-4 février 2026 : Epitech Summit 2026 - Rennes - Rennes (France) 3-4 février 2026 : Epitech Summit 2026 - Montpellier - Montpellier (France) 3-4 février 2026 : Epitech Summit 2026 - Strasbourg - Strasbourg (France) 3-4 février 2026 : Epitech Summit 2026 - Toulouse - Toulouse (France) 4-5 février 2026 : Epitech Summit 2026 - Bordeaux - Bordeaux (France) 4-5 février 2026 : Epitech Summit 2026 - Lyon - Lyon (France) 4-6 février 2026 : Epitech Summit 2026 - Nice - Nice (France) 12-13 février 2026 : Touraine Tech #26 - Tours (France) 26-27 mars 2026 : SymfonyLive Paris 2026 - Paris (France) 27-29 mars 2026 : Shift - Nantes (France) 31 mars 2026 : ParisTestConf - Paris (France) 16-17 avril 2026 : MiXiT 2026 - Lyon (France) 22-24 avril 2026 : Devoxx France 2026 - Paris (France) 23-25 avril 2026 : Devoxx Greece - Athens (Greece) 6-7 mai 2026 : Devoxx UK 2026 - London (UK) 22 mai 2026 : AFUP Day 2026 Lille - Lille (France) 22 mai 2026 : AFUP Day 2026 Paris - Paris (France) 22 mai 2026 : AFUP Day 2026 Bordeaux - Bordeaux (France) 22 mai 2026 : AFUP Day 2026 Lyon - Lyon (France) 17 juin 2026 : Devoxx Poland - Krakow (Poland) 11-12 juillet 2026 : DevLille 2026 - Lille (France) 4 septembre 2026 : JUG Summer Camp 2026 - La Rochelle (France) 17-18 septembre 2026 : API Platform Conference 2026 - Lille (France) 5-9 octobre 2026 : Devoxx Belgium - Antwerp (Belgium) Nous contacter Pour réagir à cet épisode, venez discuter sur le groupe Google https://groups.google.com/group/lescastcodeurs Contactez-nous via X/twitter https://twitter.com/lescastcodeurs ou Bluesky https://bsky.app/profile/lescastcodeurs.com Faire un crowdcast ou une crowdquestion Soutenez Les Cast Codeurs sur Patreon https://www.patreon.com/LesCastCodeurs Tous les épisodes et toutes les infos sur https://lescastcodeurs.com/

In honor of National Native American Heritage Month, SWE CEO and Executive Director Karen Horting sits down with Sarah EchoHawk, president and CEO of Advancing Indigenous People in STEM (AISES), to discuss visibility, allyship, and access for Indigenous engineers. Sarah shares her family's deep legacy of public service, the role of tribal colleges in reclaiming education, and how Indigenous knowledge systems — from fire science to environmental stewardship — can help solve global challenges. Plus, hear how employers, educators, and organizations like SWE can strengthen partnerships with AISES to ensure Indigenous voices are included in the future of STEM. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

This episode is sponsored by Resideo. In this episode, two engineers share how curiosity, empathy, and innovation shape their work creating smart home technologies. Helen Meza, systems integration lead engineer at Resideo, shares her inspiring path from Peru to the U.S. and discusses how she leads global teams with empathy and adaptability. Kyra Neal, mechanical design engineer at Resideo, reflects on how she went from designing 3D printers to designing thermostats and dehumidifiers that make tangible differences in people's lives — and how curiosity played a role in her STEM career. In conversation with Larry Guthrie, director of content strategy at SWE, hear Helen's insights on incorporating cultural awareness into leadership, Kyra's lessons on finding growth through every challenge, and how Resideo fosters an environment where engineers can make an impact. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

What is it like to navigate STEM while living with a visible or invisible disability? In this episode, Haylee Mota, a recent engineering graduate who is blind, shares how she found her path to aerospace through robotics, hands-on research, and self-advocacy — overcoming barriers like inaccessible course materials. Angelie Vincent, a mid-career aerospace engineer living with autism, diabetes, and celiac disease, reflects on lessons learned from a late diagnosis, masking in the workplace, and the power of mentorship. In conversation with Jenevieve Surkin, lead of SWE's DisAbility Inclusion Affinity Group (DIAG), hear how universities and workplaces can do better and why requesting accommodations is about creating equal access and opportunity. DIAG was formed to build a supportive, safe community for differently abled/disabled engineers, caregivers, and allies to network and develop themselves professionally. Get involved and find out about upcoming events at sites.swe.org/diag/. — The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Jessica reports LIVE from Jakarta on all the details from day two of women's podium training. World Championships Headquarters  Videos, Interviews, Podcasts, Fantasy, Guides Extended Episode + Live Q&A (Members) +30 extra minutes of analysis, behind-the-scenes secret stories, plus member questions. Here's how to ask questions live. Can't make it live? Add Club bonus episodes to your favorite podcast player (instructions here). Chapters 00:00 – Show Intro 01:02 – Zhang Qingying beam world champion prediction 03:00 – FIG Press Conference recap: AI D-scores and visa issue 08:40 – Spencer's updates: where to watch & fantasy game deadlines 11:45 – U.S. Women's Team podium training report (Josc, Skye, Dulcy, Leanne) 17:20 – Can Josc vault? Exclusive Olympic Channel interview 19:45 – Equipment update: white mats and “China mat overlay” 22:10 – Mixed Zone highlights (Malabuyo, South Africa, Asia's coach impression) 25:05 – Italy updates: Perotti, Asia D'Amato, Fioravanti AA potential 29:45 – Melnikova and Russia (AIN) podium impressions 31:30 – Flavia Saraiva's 10.0 leotard and Brazilian updates 33:10 – Funniest & coolest skills of the day (Chile, India, Portugal) 33:55 – BTS Teaser begins 34:00 – Embarrassing moments & Watanabe press conference story 36:40 – Beam fall hilarity (NZL gymnast) 38:15 – Opposite of Canadian medical intervention 40:00 – The great Indonesian tampon saga 42:25 – Sub 4: NZL, LIE, USA, CRO, BAN, GBR, POL 45:10 – Ruby Evans Amanar, GB bars, Alia Leat injury update 47:05 – Sub 5: MAS, SUI, ITA, FRA, VIE, ISL, MAR 49:00 – Thelma's floor, Osyssek's beam, Ming Van Eijken vaults 51:05 – Sub 6: AUS, EGY, BEL, LAT, ROU, MGL, SWE, CRC 53:00 – Voinea full Gothic mode, Golgota AA, Romanian updates 56:20 – Sub 7: INA, TUN, COL, PHI, MEX, SYR 58:00 – Finnegan & Malabuyo AA, Seema Tello debut 1:00:10 – Sub 8: NOR, BRA, QAT, IND, RSA, CHI 1:02:15 – Flavia & Brazil updates, Rooskrantz, Chilean grandmas 1:05:00 – Sub 9: AIN, NAM, POR, THA, BUL, SLO, CMR 1:07:25 – Melnikova Cheng, Cameroon floor joy, AIN medal watch 1:10:10 – Sub 10: ESP, AIN, HUN, HKG, CHN, KZN, CZE 1:12:25 – Zhou Yaqin & Zhang Qingying on beam, Deng Yalan vault 1:15:30 – Alba Petisco all-around standout 1:17:10 – Feedback: listener comments from Dr. Ben & Absolutely Not 1:21:20 – Show Close: Women's qualifying preview & thanks How Do I Watch the Competition? All sessions of the competition will be streamed on Eurovision Sport. Follow along here! Gymnastics Indonesia's YouTube channel will stream all qualification sessions Live scores from the FIG and Swiss Timing Check out NBC's behind-the-scenes mini-doc on the US Women's World Trials Headlines What happened at podium training today? Should we be worried about the US women? From the Olympic Channel: Joscelyn Roberson has been struggling to "find her block" on vault Skye's HUGE front-handspring front on beam Who else from Florida came to join the 2025 World Championships party? Giulia Perotti (Italy) looks ready to win all the medals Who will be the second Italian competing all-around? The D'Amato vs. Fioravanti dilemma Angelina Melnikova is so back How did her vaults look? WE NEED TO TALK ABOUT BRAZIL'S GENIUS LEOS Flavia showed beam and floor - how'd it go? Who wins the award for coolest/best/most fun skill from podium training? What were Jessica's mixed zone highlights? The FIG held a press conference today. What information did we learn? The FIG announced that "spectators will be able to see AI D-scores," but what does this mean? The FIG addressed the visa vs. FIG rules issue. What did FIG president Watanabe have to say? Jakarta Updates GymCastic Updates Subscribe to our YouTube Channel Coming Up 6 days of LIVE podcasts at World Championships in Jakarta Club members get extended coverage and can join us live to ask questions immediately after the meet Play our World Championships Fantasy Game! Win a Club Gym Nerd Scholarship: Go to our Forum > Show Stuff > GymCastic Scholarship We are matching every new sponsorship If you would like access to the club content, but aren't currently in a position to purchase a membership, all you need to do is fill out the form that's linked in our message board If you would also like to sponsor a scholarship, please email editor@gymcastic.com. Thank you! Support Our Work Club Gym Nerd: Join Here Become a Sponsor: GymCastic is matching all donations Nearly 50 scholarships have been awarded so far Learn More Headstand Game: Play Now Forum: Start Chatting Merch: Shop Now Thank you to our Sponsors Gymnastics Medicine Beam Queen Bootcamp's Overcoming Fear Workshop Resources Jakarta schedule & times: See our live podcast times on the Worlds HQ schedule Guides: Download the quick-reference guide on the Jakarta Headquarters page The Balance Beam Situation: Spencer's GIF Code of Points Gymnastics History and Code of Points Archive from Uncle Tim Kensley's men's gymnastics site Neutral Deductions   Unlock the Extended Episode Join Club Gym Nerd → Choose a plan Complete checkout — your site account is created. Log in here → /my-account/ Return to this page and refresh. The extended player appears automatically.

This episode is sponsored by Draper. How do you know when it's the right moment to make a career change? In this episode, Maureen Masiulis, director of space sensing, domain awareness, and control at Draper and acting vice president and general manager of Draper's Space Systems Business Unit, shares her STEM journey from robotics in middle school to leadership roles in the aerospace and defense industries. In conversation with host Sam East, she reflects on the pivotal moments when she had to decide whether to move on and details how her volunteer experiences within SWE shaped her leadership skills. Hear how Maureen balances career growth with family and personal priorities and why Draper's nonprofit model and programs like Draper Scholars make it a unique place for engineers to grow their careers. ----- The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Join us for a thrilling episode as we sit down with Joshua Amen, a seasoned ghost hunter with over 12 years of spine-chilling paranormal investigations. Joshua has encountered everything from full-body apparitions to inexplicable phenomena that defy explanation. With a scientific approach, Joshua and his team explore whether the tools used in paranormal investigations could influence entities from beyond the veil. Could black light be the key to dispersing malevolent spirits? Joshua also shares a jaw-dropping encounter with a mysterious creature—possibly a werewolf or dogman—that challenges our understanding of the unknown. Dive into his riveting experiences, detailed in his compelling book, Tales from a Ghosthunter: The Collective Stories. https://popl.co/card/6RfQR13p/2/sWe are thrilled to announce the official launch of Let's Get Freaky merchandise! Our collection includes hoodies, t-shirts, mugs, stickers, and more. Explore the full range at http://tee.pub/lic/aQprv54kktw.Do you have a paranormal or extraordinary experience to share? We'd love to hear from you! Contact us to be a guest on the Let's Get Freaky podcast. Email us at letsgetfreakypodcast@mail.com or reach out via social media on Facebook, Instagram, X, TikTok, or YouTube at @tcletsgetfreakypodcast. Connect with us at https://linktr.ee/letsgetfreaky.

Gäster: Agnes Matsdotter, Clara Kristiansen, Isidor Olsbjörk För 90SEK/mån får du 5 avsnitt i veckan:4 Vanliga AMK MORGON + AMK FREDAG med Isak Wahlberg Se till att bli Patron via webben och inte direkt i iPhones Patreon-app för att undvika Apples extraavgifter:Öppna istället din browser och gå till www.patreon.com/amkmorgon Gå och kolla på Clara och Marcus föreställning "Nära Vänner"https://gamang.se/nara-vanner/ Pre-savea Agnes nya låt "Comme ci, comme ça":https://link.ninetone.com/r1qdx6pRelevanta länkar: …Falungonghttps://sv.wikipedia.org/wiki/Falungong …kalsongreklamenhttps://static.bonniernews.se/ba/236a2d88-9897-440d-ad01-c668a10667c7.jpeg?crop=889%2C500%2Cx56%2Cy0&width=1920&auto=webp&enable=upscale …fildrinken på röda husethttps://rodahuset.nu/wp-content/uploads/2025/08/Cocktail-experience_WEB_13aug2025-SWE.pdf …Daniela Rathanahttps://www.instagram.com/danielarathana/ …Yung Leanhttps://sv.wikipedia.org/wiki/Yung_Lean…Östermalmskrogenhttps://www.svt.se/nyheter/lokalt/stockholm/tre-gaster-sjuka-i-sepsis-lyxkrog-pa-ostermalm-stangerLåtarna som spelades var:En natt i Köpenhamn - BAO, Helen Sjöholm, Tommy KörbergJumpin' Jack Flash - The Rolling Stones Alla låtar finns i AMK Morgons spellista här:https://open.spotify.com/user/amk.morgon/playlist/6V9bgWnHJMh9c4iVHncF9j?si=so0WKn7sSpyufjg3olHYmg

Suzanna Valdez Wolfe, CEO of the Society of Hispanic Professional Engineers (SHPE), sits down with Karen Horting, CEO and executive director of the Society of Women Engineers (SWE) in honor of Hispanic Heritage Month. Suzanna shares her journey from El Paso, Texas, to leadership roles in Washington, D.C., and now at SHPE, where she champions familia, community, and belonging in STEM. Hear the unique challenges and opportunities facing Latinas in engineering, the importance of mentorship and leadership development, and how SHPE and SWE are working together to strengthen the STEM pipeline. Plus, explore how advocacy at the local and national levels ensures Hispanic voices are represented in shaping the future of STEM. ----- The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Ken Rosenthal's Bad LookLetter From Beyond The GraveHe Must Be Really Good!TEMU For QB'sWe're Not Doing Stonehenge ToniteWe Talkin' About Practice!About That Injury Report ThingHerm Edwards, Your Rant Is SafeAnother Reason To Loathe The ChiefsA Day Later, SorryDo You Even Kicker, Bro?That Play You Saw? No You Didn'tJoe Burrow To The SlaughterGo To the Game, It'll Be Fun!Cracker Barrell Total Surrender!Our Sponsors:* Check out Hims: https://hims.com/CZABE* Check out Indeed: https://indeed.com/CZABEAdvertising Inquiries: https://redcircle.com/brandsPrivacy & Opt-Out: https://redcircle.com/privacy

Kelsea Fitzgerald, director of operations at Chevron's Engineering and Innovation Excellence Center (ENGINE), shares her journey from studying chemical engineering in the U.S. to taking on a global leadership role in Bengaluru, India. In conversation with Larry Guthrie, director of content strategy at SWE, Kelsea encourages women and allies to build self-trust by taking uncomfortable steps out of their comfort zones. Plus, hear what Kelsea loved the most about WE Local Bengaluru, including moderating a panel on self-trust and finding a sense of community with new STEM connections. WE Local conferences events bring together engineers and technologists of all genders for professional development sessions, networking opportunities, and inspirational speakers. Find the next WE Local conference near you at welocal.swe.org ----- The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Join us in this exciting episode of The Edge of Show as we dive deep into the world of Bitcoin DeFi with two industry leaders: MacLane Wilkison, co-founder and CEO of Threshold Labs, and Jameel Khalfan, head of ecosystem development at Sui Foundation.In this episode, we explore:The transformative potential of TBDC (Threshold Bitcoin Decentralized Currency) and how it is powering Bitcoin's breakout moment on the SWE ecosystem.The rapid evolution of Bitcoin DeFi and the seamless bridging and yield strategies that are emerging.Insights into the future of Bitcoin as a dynamic financial asset and its role in the broader DeFi landscape.Discover how the integration of Bitcoin with decentralized finance is reshaping the financial ecosystem, making it more accessible and user-friendly. We also discuss the importance of security, liquidity, and the collaborative efforts between Threshold and SWE to foster innovation and growth in the Bitcoin space.Whether you're a seasoned crypto enthusiast or just starting your journey, this episode is packed with valuable insights and exciting developments in the world of Bitcoin and DeFi.Don't forget to like, subscribe, and hit the notification bell to stay updated on our latest episodes!Support us through our Sponsors! ☕

This episode is sponsored by Jabil. Cassie Gruber, director of sustainability business solutions at Jabil, explores how engineers can shape a more sustainable future in this episode of Diverse. In conversation with Larry Guthrie, director of content strategy at SWE, Cassie explores her unconventional career path from studying art and philosophy to leading global sustainability strategies in manufacturing. Hear the biggest sustainability challenges she is seeing across 16 sectors, the importance of collaboration in the circular economy, and why Jabil is a unique place for engineers to grow and thrive in their careers. ----- The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

• Crustacean talk kicks things off with crawfish boils described as eating cockroaches from ditches while Ross discovers Virginia Beach crab cakes the size of baseballs with zero filler that make landlocked Chicagoland versions look like sad breadcrumb sandwiches• Colton finally pulls the trigger on shop electricity after writing Centerpoint a check though the five to eight week wait has him considering temporary power solutions while exploring contractor work with SWE homes and reconnecting with Rick who just dropped thirteen grand on a laser cleaning machine that can strip paint off drywall across the room• Poison ivy becomes the episode villain attacking Colton in places that shall remain nameless while he channels his grief into songwriting for his grandfather's funeral using Zach Bryan melodies to capture fishing memories and family time themes• Jess returns from New Jersey family time with shirt printing machine connections before discovering his concrete removal nightmare doubled when plumbers demanded another trench for proper drainage while juggling work from home customers who freak out about power outages during electrical upgrades• Built-in cabinet designs meet reality when expensive plantation shutters don't fit the new configuration forcing difficult conversations about seven thousand dollar window treatments versus custom cabinetry dreams while Ashley handles the growing bid workload with professional scripts• Trailer organization becomes an art form as Jess transforms his twenty-four foot beast into a mobile workshop complete with electric jack systems and custom shelving that required cutting sixty-seven degree angles his chop saw couldn't handle until he got creative with wedge blocks• Ross tackles rolling jewelry cabinet construction with locking mechanisms and removable glass tops while figuring out battery-powered LED lighting systems that need magnetic mounting solutions for portable display cases that can roll around retail spaces• Work truck history spans from horse-drawn wagons to Model T runabouts revealing that pickup truck terminology originated in nineteen twenty-five when Ford first offered factory-installed beds instead of chassis-only sales requiring custom carpenter-built wooden cargo areas• Electric vehicle debates explore charging infrastructure challenges and carbon capture gasoline production while non-alcoholic beer production gets explained through fermentation timing and dilution techniques for those maintaining sobriety goals during business building phases• Workshop nuggets include motion-sensor LED strips that charge via USB-C for trailer lighting systems plus Kreg jig depth adjustments for thinner plywood that prevents screw blowout while LED strip connectors save the day when expensive light installations get damaged during installation

In honor of the 10th anniversary of SWENext, this episode of Diverse brings a special mother-daughter conversation. Bharti B., operations manager at Keysight Technologies, reflects on what it was like to be one of the few women in her computer science classes growing up. Aavya B., who is a SWENext influencer as well as Bharti's daughter, shares how she's carrying that legacy forward by founding the first SWENext Club in India and pursuing her dream of becoming a biomedical engineer. In conversation with Sam East, Bharti and Aavya discuss how opportunities for girls in engineering have evolved, the power of role models, and the unique bond that comes from sharing a STEM passion across generations. Know a precollege kid interested in STEM? Join SWENext, the Society's free program that allows students ages 5 through 18 to join the SWE engineering and technology community! Learn more and get involved at swe.org/outreach/swenext/ --- The Society of Women Engineers is a powerful, global force uniting 50,000 members of all genders spanning 85 countries. We are the world's largest advocate and catalyst for change for women in engineering and technology. To join and access all the exclusive benefits to elevate your professional journey, visit membership.swe.org.

Our 216th episode with a summary and discussion of last week's big AI news! Recorded on 07/11/2025 Hosted by Andrey Kurenkov and Jeremie Harris. Feel free to email us your questions and feedback at contact@lastweekinai.com and/or hello@gladstone.ai Read out our text newsletter and comment on the podcast at https://lastweekin.ai/. In this episode: xAI launches Grok 4 with breakthrough performance across benchmarks, becoming the first true frontier model outside established labs, alongside a $300/month subscription tier Grok's alignment challenges emerge with antisemitic responses, highlighting the difficulty of steering models toward "truth-seeking" without harmful biases Perplexity and OpenAI launch AI-powered browsers to compete with Google Chrome, signaling a major shift in how users interact with AI systems Meta study reveals AI tools actually slow down experienced developers by 20% on complex tasks, contradicting expectations and anecdotal reports of productivity gains Timestamps + Links: (00:00:10) Intro / Banter (00:01:02) News Preview       Tools & Apps (00:01:59) Elon Musk's xAI launches Grok 4 alongside a $300 monthly subscription | TechCrunch (00:15:28) Elon Musk's AI chatbot is suddenly posting antisemitic tropes (00:29:52) Perplexity launches Comet, an AI-powered web browser | TechCrunch (00:32:54) OpenAI is reportedly releasing an AI browser in the coming weeks | TechCrunch (00:33:27) Replit Launches New Feature for its Agent, CEO Calls it ‘Deep Research for Coding' (00:34:40) Cursor launches a web app to manage AI coding agents (00:36:07) Cursor apologizes for unclear pricing changes that upset users | TechCrunch       Applications & Business (00:39:10) Lovable on track to raise $150M at $2B valuation (00:41:11) Amazon built a massive AI supercluster for Anthropic called Project Rainier – here's what we know so far (00:46:35) Elon Musk confirms xAI is buying an overseas power plant and shipping the whole thing to the U.S. to power its new data center — 1 million AI GPUs and up to 2 Gigawatts of power under one roof, equivalent to powering 1.9 million homes (00:48:16) Microsoft's own AI chip delayed six months in major setback — in-house chip now reportedly expected in 2026, but won't hold a candle to Nvidia Blackwell (00:49:54) Ilya Sutskever becomes CEO of Safe Superintelligence after Meta poached Daniel Gross (00:52:46) OpenAI's Stock Compensation Reflect Steep Costs of Talent Wars       Projects & Open Source (00:58:04) Hugging Face Releases SmolLM3: A 3B Long-Context, Multilingual Reasoning Model - MarkTechPost (00:58:33) Kimi K2: Open Agentic Intelligence (00:58:59) Kyutai Releases 2B Parameter Streaming Text-to-Speech TTS with 220ms Latency and 2.5M Hours of Training       Research & Advancements (01:02:14) Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning (01:07:58) Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity (01:13:03) Mitigating Goal Misgeneralization with Minimax Regret (01:17:01) Correlated Errors in Large Language Models (01:20:31) What skills does SWE-bench Verified evaluate?       Policy & Safety (01:22:53) Evaluating Frontier Models for Stealth and Situational Awareness (01:25:49) When Chain of Thought is Necessary, Language Models Struggle to Evade Monitors (01:30:09) Why Do Some Language Models Fake Alignment While Others Don't? (01:34:35) Positive review only': Researchers hide AI prompts in papers (01:35:40) Google faces EU antitrust complaint over AI Overviews (01:36:41) The transfer of user data by DeepSeek to China is unlawful': Germany calls for Google and Apple to remove the AI app from their stores (01:37:30) Virology Capabilities Test (VCT): A Multimodal Virology Q&A Benchmark

You've heard this one before: the sexiest quality in the world is confidence. And that's true...but...it's not always easy to muster. Confidence is an inside job, but here's the good news. Once you generate it on your own terms, life - and sex - gets a whole lot easier. So on today's Ask Emily show, I'm joined by my dear friend and former co-host Menace from The Woody Show, as we talk confidence, jealousy, What happens when you adore your partner, but can't stop thinking about their high body count? Or want to have hot phone sex, but don't know where to start? We talk through both. How about when you feel too short to find a great partner, or don't know how to be dominant in bed? We help you think through both. Finally: discovering your partner's can of delay spray. Our listener wonders whether he's sleeping with someone else; we think it's time for open communication. For their buy 1 get 1 50% off deal, head to 3DayBlinds.com/SWE.   .    Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com

We're back for another AITA episode! This time it's all about sexual mismatches. First: they've got a 12-year age difference, and she suspects his sex drive is dwindling. She doesn't want to cheat, but she also wants more sex…what now? Next, he wants sex for pleasure, she was raised to do it only for procreation. Is there any way to meet in the middle? A self-professed "pillow princess" writes in, is she the asshole for wanting to receive pleasure – but not give it? Finally, they're in a dead bedroom of 10 years. “She's repulsed by my touch” he writes, but is he really asserting himself to make healthy changes? You decide For their buy 1 get 1 50% off deal, head to 3DayBlinds.com/SWE.   .    Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com

Ever wondered: Am I the asshole for watching porn without my partner knowing? Am I the asshole for feeling a little misled about our sex life before we walked down the aisle? Or how about: Am I the asshole for just feeling plain bored with our sex life? You're not alone. Today, Producer Erica and I read your AITA questions and let you know if you're approaching asshole territory. In today's episode, you'll learn: How to be a supportive partner to someone with sexual trauma How to get clearer on your sexual desires About the experiences of sex workers Show Notes: Try OSEA Today! Get 10% off your first order sitewide with code SWE at OSEAMalibu.com Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com

The Sex With Emily Hotline is open! On this episode, I'm joined by my Producer, Erica, as we talk about disappearing erections, prostate massages, and so much more.  First, what do you do when your partner has health issues that cause them to lose their erections? I give one caller ideas for how to work around this issue and thoughts on what might be causing it in the first place. Next, why don't couples fantasize together more often? A couple calls in to share their sexual fantasies, and we give you inspo on how to do the same with your partner. When your partner wants a prostate massage, how do you pull it off? I tell you what to feel for, how to apply pressure and fun toys you can use. Finally, when your partner has never had an orgasm before, how can you help? And should you? In this episode, you'll learn: Why erectile challenges aren't just about the penis—and how you can spice things up while finding solutions How to co-create fantasies with your partner for hotter, more connected sex The secret to unlocking prostate pleasure (and why so many men are missing out) Show Notes: Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Head to MarleySpoon.com/offer/swe and use code SWE for up to 27 FREE meals! Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com

Two of your most powerful sex accessories? Your hands. Hands set the tone of your sexual energy. For example, caressing their cheek while you make out versus pinning their hands down while you have sex. And while we talk a lot about what to do with our mouths or genitals on this show, today I'm focusing on a lost art: hand play. Specifically, how to finger, give a hand job, and use your hands with sexual intention.  Today, my Producer, Erica, and I first share how to penetrate a vulva with your fingers. We give you tricks to stimulate the labia and clitoris, different forms of pressure and touch, and how to find the G-spot when you finger. Next, we give the penis some love with hand job techniques and upgrades like toy play and perineum stimulation. Finally, we discuss secondary erogenous zones and answer your hand play questions. In this episode, you'll learn: Why hand play—fingering and handjobs—is the underrated pleasure tool you need to master. How to turn up the heat with expert hand techniques for both vulvas and penises. The secret to making manual stimulation feel even better than penetrative sex. Show Notes: Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Try Nutrafol Today! Head to Nutrafol.com and use code "SWE" at checkout for $10 off your first month's subscription and free shipping. Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com

Dating as an adult is exhausting. While experience comes with age, actually applying that experience is easier said than done…even for professionals.  With me today are two therapy professionals who also happen to be partners: John Kim and Vanessa Bennett. Their book “It's Not Me, It's You” offers an anecdotal road map of how to “relationship better,” from overcoming commitment issues, to fanning the sex flames, to breaking the blame cycle. John and Vanessa expertly dissect their own relationship to inspire us to create healthy bonds with the people we love. Listen to learn why “the one” is a myth, how to wean ourselves off codependency and how to replace obligatory sex with erotic collaboration. In this episode, you'll learn: Why “The One” is a myth and what actually makes relationships last How to turn obligatory sex into erotic collaboration The secret to avoiding resentment before it ruins your connection Show Notes: More John Kim: Instagram | Website | Book More Vanessa Bennett: Instagram | Website | Book How to Have the Best Sex of Your Life in 2025- Sign up for my FREE masterclass today! For their buy 1 get 1 50% off deal, head to 3DayBlinds.com/SWE.   Join the SmartSX Membership: Access exclusive sex coaching, live expert sessions, community building, and tools to enhance your pleasure and relationships with Dr. Emily Morse. Yes! No! Maybe? List & Other Sex With Emily Guides: Explore pleasure, deepen connections, and enhance intimacy using these Sex With Emily downloadable guides.  SHOP WITH EMILY! (free shipping on orders over $99) The only sex book you'll ever need: Smart Sex: How to Boost Your Sex IQ and Own Your Pleasure Want more? Visit the Sex With Emily Website  Let's get social: Instagram | X | Facebook | TikTok | Threads | YouTube Let's text: Sign up here Want me to slide into your email inbox? Sign Up Here for sex tips on the regular. See the full show notes at sexwithemily.com