Beyond the Hype

In this episode, Oliver Cronk is joined by colleagues David Rees, Hélène Sauvé, Ivan Mladjenovic and Emma Pearce. Together, they delve into the practical applications and limitations of agentic AI and its implications for enterprise AI deployments. The team shares insights from the ‘Infer' research and development projects, through which Scott Logic produced and open-sourced InferLLM (a local, personalised AI agent) and InferESG (which uses AI agents to identify greenwashing in Environmental, Social and Governance reports). With real-world examples and expert perspectives, the panel provides a nuanced view of whether fully autonomous agents are hype or reality in 2025. They discuss the balance between human oversight and automation, and emphasise the importance of transparency and traceability in AI systems. They also consider the ethical considerations of self-building agents and the challenges of ensuring reliable AI outputs. Have a listen to gain a deeper understanding of the evolving landscape of agentic AI and its potential impact on various sectors. Useful links for this episode InferLLM on GitHub – Open-sourced by Scott Logic InferESG on GitHub – Open-sourced by Scott Logic InferESG: Augmenting ESG Analysis with Generative AI – David Rees, Scott Logic InferESG: Finding the Right Architecture for AI-Powered ESG Analysis – David Rees, Scott Logic InferESG: Harnessing agentic AI for due diligence – Scott Logic case study Beyond the Hype: Will we ever be able to secure GenAI? – Scott Logic Beyond the Hype: Is architecture for AI even necessary? – Scott Logic Draft classification for different types of Enterprise AI deployment – Oliver Cronk, Scott Logic

In this episode, Oliver Cronk is joined by colleagues Jess McEvoy and James Heward, and Atom Bank's Head of AI and Data Science, Russell Collingham, to tackle the provocative question: "Is architecture for AI even necessary?" They explore the transformative impact of generative AI and the critical role of architecture in ensuring sustainable and scalable implementations. Russell shares Atom Bank's approach to building reusable AI components while navigating security and governance challenges. James emphasises the evolving role of architects in balancing innovation with practical constraints, arguing that architecture remains essential for mission-critical applications. Jess provides a public sector perspective, highlighting the need for a balanced approach to AI regulation and innovation. The episode also delves into the geopolitical implications of AI, the importance of data sovereignty, and the potential for the UK to develop its own AI strategy. Have a listen to gain insights into the ongoing debate about the necessity of architecture in the AI era and the strategic considerations for leveraging AI effectively. Useful links for this episode There is more than one way to do GenAI – Oliver Cronk, Scott Logic Are we sleepwalking into tomorrow's AI challenges? – Oliver Cronk, Scott Logic Building Effective AI Agents – Anthropic Fully Autonomous AI Agents Should Not Be Developed (PDF) – Hugging Face Will the US cloud soon be illegal in the EU? – CIO

In this episode of Beyond the Hype, Oliver Cronk is joined by Bradon Rogers from Island, along with Scott Logic colleagues Dean Kerr and Robat Williams, to explore the potential of enterprise browsers. They delve into the advantages of enterprise browsers over standard options like Chrome and Edge, particularly in terms of security and productivity. Bradon describes how enterprise browsers, built on a Chromium foundation, offer a familiar user experience while integrating robust security features and application delivery capabilities. The discussion also covers the challenges of using consumer-grade browsers in corporate environments, the benefits of desktop container technologies, and the potential for enterprise browsers to simplify security and enhance user productivity. Join the panel as they consider whether enterprise browsers will transform the way organisations manage and secure their digital workspaces. Useful links for this episode Emerging Tech: Security — The Future of Enterprise Browsers – Gartner Web browsers: Reimagining remote work needs at the enterprise level – CIO The rise of the enterprise browser and what's next for secure browsing – VentureBeat

In this episode, Oliver Cronk is joined by Technology Lead Andrew Carr and CTO Colin Eberhardt to delve into the evolving nature of technology strategy within organisations. As technological advancements accelerate, they question the relevance of a traditional long-term technology strategy and whether it has become an industry buzzword in itself. They explore the annual ritual of tech predictions and strategic planning, and whether it is practical or performative. They also debate the balance between addressing technical debt and fostering innovation, the influence of analyst-driven hype, and the critical decision of buy versus build. There's a lot of hype to navigate around, but they bring their range of experiences to bear in providing a nuanced perspective on how to create effective technology strategies that genuinely add value – by aligning technology strategies with business goals. Links from this episode How to build an effective technical strategy – LeadDev Better than a secret weapon: 5 steps to a winning technology skills strategy  – TechUK The Tech:Forward recipe for a successful technology transformation – McKinsey

In this episode, Oliver Cronk is joined for a discussion on observability by Scott Rowan, Senior Developer at Scott Logic, and Daniel Gomez Blanco, Principal Engineer at Skyscanner and a member of the Open Telemetry Governance Committee.   The conversation explores what observability means in modern distributed software architectures, how it differs from traditional monitoring, and the challenges of implementing observability at scale. The discussion touches on practical aspects of implementing observability and how this approach can lead to faster problem detection and resolution, as well as cost savings by reducing the volume of less useful data collected.   The rise in popularity of observability has gone hand in hand with the rise of microservices and event-driven architectures. But although it's a relatively new kid on the block, is observability hyped or a necessary evolution in managing modern software systems? Links from this episode What is Observability? – The New Relic perspective What is Observability? – The Dynatrace perspective OpenTelemetry website OpenTelemetry introduction at CloudNativeCon

In this episode, Oliver Cronk, Doro Hinrichs and Kira Clark from Scott Logic are joined by Peter Gostev, Head of AI at Moonpig. Together, they explore whether we can ever really trust and secure Generative AI (GenAI), while sharing stories from the front line about getting to grips with this rapidly evolving technology. With its human-like, non-deterministic nature, GenAI frustrates traditional pass/fail approaches to software testing. The panellists explore ways to tackle this, and discuss Scott Logic's Spy Logic project which helps development teams investigate defensive measures against prompt injection attacks on a Large Language Model. Looking to the future, they ask whether risk mitigation measures will ever be effective – and what impact this will have on product and service design – before offering pragmatic advice on what organisations can do to navigate this terrain. Links from this episode Prompt injection explained, with video, slides, and a transcript – Simon Willison's Weblog Spy Logic on GitHub Spy Logic – Doro Hinrichs and Heather Logan How the tables turned – My life with Spy Logic – Kira Clark

In this episode, Oliver Cronk and David Rees from Scott Logic are joined by Hannah Smith, Director of Operations at Green Web Foundation, an organisation aiming to make the internet fossil-free by 2030. Together, they explore the potential benefits and limitations of ‘carbon aware' computing, which involves scheduling computational workloads during times or in locations where energy sources have lower carbon emissions. With the simplicity of ‘carbon aware' being at odds with the complexities of managing the electricity grid, they move on to discussing ‘grid aware'. This concept goes beyond carbon intensity to take into account factors like grid stability, demand patterns, and energy sources being curtailed. Ultimately, they consider just how marginal the gains are likely to be. With global emissions from cloud computing already exceeding those from commercial flights, is ‘carbon aware' a distraction from getting to grips with the scale of the challenge we face? Links from this episode Green Web Foundation What software engineers need to know about how energy grids work - Green Web Foundation (thegreenwebfoundation.org) Full Grid Aware Research on GitHub Sustainability topic on the Scott Logic Blog

In this episode, Colin Eberhardt is joined by Jess McEvoy and Peter Chamberlin, who have both spent many years in senior roles within public sector organisations. Their conversation covers the excitement and concerns around AI, both from a citizen's perspective and for those building public services. They discuss the UK government's approach to addressing AI challenges with its pro-innovation mantra, and whether this creates the right environment for success. They also touch on the need for global cooperation when tackling what are increasingly becoming global challenges. Links from this episode A guide to using artificial intelligence in the public sector  National AI Strategy (PDF) A pro-innovation approach to AI Prime Minister's speech on AI: 26 October 2023 The Bletchley Declaration – November 2023 Policy implications of artificial intelligence (AI)  A pro-innovation approach to AI regulation: government response  AI Safety Institute - Third progress report

In this episode, Oliver Cronk, Andrew Carr and David Hope talk about the ever-changing world of data, with conversations moving from data warehouse to data lake, and data mesh to data fabric. They discuss the importance of data ownership and common tooling, and their view that data mesh is an approach rather than an architecture. Links from the podcast Data Mesh by Zhamak Dehghani Data Fabric as Modern Data Architecture by Alice LaPlante Principles of Data Fabric by Sonia Mezzetta

In this episode, Oliver Cronk is joined by experts including Jeremy Axe, Group CTO at DS Smith, and consultants Darren Smith and Katie Davis from Scott Logic. Together, they unpack topics like the energy usage and carbon emissions of IT infrastructure, the challenges in accurately measuring sustainability, and whether claims of ‘green tech' are substantiated or just hype. The conversation covers the nuances around operational versus embodied carbon emissions, the sustainability trade-offs involved in cloud versus on-premises hosting, and the need for transparency and standards around eco-friendly tech claims. Links from the podcast Proposed Technology Carbon Standard – open sourced by Scott Logic under a CC BY-NC-SA 4.0 licence Scott Logic blog posts on Sustainability – including the Conscientious Computing series DS Smith's approach to sustainability Scott Logic's approach to supporting technology sustainability

In this episode, Colin is joined by Rebecca Rumbul, CEO of Rust Foundation, and Mirko Boehm from Linux Foundation Europe. Between them, they have decades of experience in open source. They start by discussing the critical role open source has grown to play in the world of software and how this, along with its growing complexity, presents significant challenges. They then turn their attention to the Cyber Resilience Act (CRA), a piece of EU legislation that is actively under development, designed to make end-user products more secure. Early drafts of this act detailed significant obligations on open source maintainers, despite the fact that they often work without financial reward. This caused concern, fear and some anger in the open source community. They discuss the latest update to the CRA, which has thankfully addressed these concerns, and ponder whether it will actually solve the problems it has set out to tackle. Links from the podcast: Panel Discussion: The Impact of the CRA on the Open Source Ecosystem – Cheukting, Mirko & Greg, Laura, Justin, Philip The EU's new Cyber Resilience Act is about to tell us how to code – Bert Hubert's writings Will the Cyber Resilience Act help the European ICT sector compete? Understanding the Cyber Resilience Act: What Everyone involved in Open Source Development Should Know EU CRA: What does it mean for open source? – Bert Hubert's writings  The EU's Proposed CRA Law May Have Unintended Consequences for the Python Ecosystem

In this episode, Oliver and Peter from Scott Logic are joined by Laura Bell Main, CEO and co-founder of SafeStack, for a lively discussion on DevSecOps. They touch on techniques such as “assume breach” and “shift left”, and the relationship to DevOps, which is trying to solve a similar problem. However, much of the discussion focused quite simply on security and the challenges this presents, especially when this so often falls to a very small team or an individual. Laura mentioned OneHourAppSec at the end of the recording. It's a free appsec program that helps software teams worldwide bring security to their software development lifecycle in just one hour per sprint. You can join at any time, from anywhere. She'd love you to check it out. Links from the podcast: The Assume Breach Paradigm  Shift left vs shift right: A DevOps mystery solved  

In this episode, Colin Eberhardt is joined by Bailey Hayes from Cosmonic and Sean Isom from Adobe – between them, they have a wealth of WebAssembly experience and insights. They start by discussing the early days of WebAssembly, and where it all began, in the browser. They debate the infamous tweet that compares WebAssembly to Docker, and look at the future promise of the Component Model, which aims to break down language barriers. Given the amount of activity happening in the WebAssembly/WASI/Component Model space, there are a few upcoming events you might be interested in: WasmCon, 6-7 Sept Componentize the World Hackathon, 8 Sept Enjoy this episode! Links from the podcast: WebAssembly nanoprocesses: making WebAssembly outside-the-browser secure by default AutoCAD featured in Google I/O Keynote WASI - WebAssembly System Interface How Prime Video updates its app for more than 8,000 device types Introducing the Disney+ Application Development Kit (ADK) Bytecode Alliance If WASM+WASI existed in 2008, we wouldn't have needed to created Docker WebAssembly Component Model Photoshop's journey to the web WebAssembly: An Updated Roadmap for Developers  

In this episode, Colin Eberhardt is joined by colleagues Oliver Cronk, Chris Price and James Heward for a lively debate on whether the latest advances in generative AI are going to threaten our jobs – are we going to be made redundant by our own creation? The discussion starts with a quick summary of the latest advances in AI, and considers the nascent reasoning capabilities these models exhibit. The discussion then explores the various tasks and roles required to build software applications, and considers the impact this technology will have. Links from the podcast: AI is morphing from tool to platform (and the next technology epoch begins) – some personal reflections on the impact of this technology Catching up with OpenAI – a quick update on the most recent advances and research from OpenAI An OpenAI API primer – a three-part series that covers explores the API, explaining concepts such as few-shot learning Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI – Lex Fridman and Sam Altman chat about the future of AI Sparks of Artificial General Intelligence: Early experiments with GPT-4 – Researchers find that GPT4 is more than just a text generation engine

In this episode – the second of a two-parter – Oliver Cronk and Colin Eberhardt talk to Denis Mandich, CTO of Qrypt, a company that creates quantum-secure encryption products. Their conversation covers the perils of bad random number generation, which undermines our security protocols, and the growing threat that Quantum Computers will ultimately render our current cryptographic techniques useless – an event dubbed ‘Y2Q', in a nod to the Y2K issue we faced over twenty years ago. Links from the podcast: Qrypt – the company where Denis is CTO A 'Blockchain Bandit' Is Guessing Private Keys and Scoring Millions Y2Q: quantum computing and the end of internet security

In this episode – the first of a two-parter – Colin Eberhardt and Oliver Cronk talk to Denis Mandich, CTO of Qrypt, a company that creates quantum-secure encryption products. Denis is also a physicist and has a wealth of knowledge when it comes to Quantum Computing.  Their conversation covers the physics that underpins this technology, including the strange concepts of superposition and entanglement. They cover the practical aspects of cloud-based Quantum Computers and what can be achieved today through circuit design. Finally, they look to the future to discuss the impact this breakthrough technology is likely to have. Links from the podcast: Qrypt – the company where Denis is CTO Quantum Fields: The Real Building Blocks of the Universe – a fantastic lecture from the Royal Institution that gives a detailed, yet accessible, introduction to Quantum physics Einstein's "Spooky Action at a Distance" Paradox Older Than Thought – some good background on the Einstein quote Qiskit – IBM's open source toolkit for creating quantum circuits

In this episode, Colin Eberhardt is joined by his colleague, Oliver Cronk, and Chris Booth from NatWest for a lively discussion about the much-hyped ChatGPT. They discuss the origins of this technology and the recent advances that contributed to its success. From there, they look at the new discipline of prompt engineering, and how this approach has democratised access to cutting-edge AI. They also consider the emotive response they feel towards this technology, and the shared belief that this is going to be huge! As this podcast aims to look ‘beyond the hype', they also talk about practical applications and the limitations of this technology. Links from the podcast: ELIZA - one of the earliest attempts at creating a conversational AI Attention Is All You Need - the paper that describes the GPT-series architecture  An OpenAI API primer - a three-part series that covers explores the API, explaining concepts such as few-shot learning Prompt Engineering GitHub Copilot – an AI programming pair, based on a large language model WebGPT - Improving the Factual Accuracy of Language Models through Web Browsing

In this episode, Colin Eberhardt is joined by colleagues Oliver Cronk, Peter Chamberlin and Chris Price for a lively discussion about blockchain. They start by looking at the mechanics of bitcoin and the economic incentive model formed by proof of work consensus. From there, they discuss enterprise or permission blockchain, which leads them to discuss some specific use cases – for example, the oil market supply-chain challenges. They also discuss technologies which are blockchain-like, but prefer not to use that term. Finally, they ask themselves the question, is blockchain just hype? Spoiler alert: for the most part the answer is yes – this is hype! Links from the podcast: Bitcoin: A Peer-to-Peer Electronic Cash System – Satoshi Nakamoto The DAO Video: Streamlining physical energy post-trade processes – Etienne Amic (CEO Vakt.io) Building the first enterprise blockchain platform – Lucy Kurian & Shodhan Sheth Certificate Transparency

In this episode, Colin Eberhardt is joined for a discussion about Web3 by his colleague Oliver Cronk, and guests Johanna Eiramo from the Finnish National Gallery, and Lilly Pencheva, a Blockchain & Web3 Specialist. As it's quite a challenging and volatile topic, we should state that the opinions raised in this podcast are personal views rather than the views of any current or former employer. In the discussion, they ask the question “What is Web3?” and explore what it means to be a decentralised technology. They discuss the metaverse, Roblox and other virtual environments, and very briefly touch on Non-Fungible Tokens (NFTs). But for the most part, they talk about exploration and innovation. They begin the discussion with Johanna talking about bringing digital artwork into the virtual world of the metaverse, and the Finnish National Gallery's motivation for experimenting with technology. Links from the podcast: Our (museum) eyes are focused on the future The Finnish Metagallery Decentraland What is Web3 and why is it important? The Ethereum Merge Can Web3 beat public cloud?

In this episode, Simon Martin, Chris Price and Rob Pilling share their interest and insights into Rust. This relatively new programming language has caught the attention of the development community, being voted the ‘most-loved' language seven years in a row in the StackOverflow survey. They discuss a wide range of topics, including what it means to be a system-programming language, and what brought each of them to Rust. They ponder what it is that makes a programming language successful and, given this, consider what the future might hold for Rust. We pick up the conversation where Chris is trying to describe the nature of Rust through a bicycle metaphor. Yes, really… and we find out later in the podcast that this metaphor works surprisingly well. Links from the podcast: Nerd Sniping - xkcd webcomic StackOverflow 2022 Survey - most-loved language WebAssembly Asynchronous Programming in Rust

In this month's episode, we tackle a wide range of topics relating to the role of technology innovation, and the perils of hyped technology, within Digital Government. Colin is joined by colleagues Jess McEvoy and Alex Segrove, who have both spent many years working in technology roles in the UK Government. They discuss how citizen-facing services have some challenging technology constraints simply because they must be usable by everyone.  We also touch on the fascinating dynamic that exists between the government and technology vendors and partners. These suppliers have a lot of expertise and IP to offer that can solve some of the gnarly problems that exist on government-scale systems. However, care must be taken to avoid outsourcing your entire technology strategy. It is vital that the government remains an intelligent buyer. Finally, we consider the sheer scale of government spending, £3.7 billion per annum, and the positive impact this has through stimulating innovation and creating opportunities. This investment has contributed to the UK Government's success in claiming the number 2 spot in the OECD Digital Government Index. Links from the podcast: G-Cloud & DOS Spending Review to December 2021 – Advice Cloud GOV.UK – The best place to find government services and information  UK claims number 2 spot in OECD digital government rankings – Jess McEvoy

In this month's episode we talk about Behaviour Driven Development (BDD), a testing practice where system behaviours are captured in a human readable Domain Specific Language (DSL), which are automated and executed. Colin is joined by colleagues Daniel McNamee, Xin Chen and Jack Arnstein who bring their many years of testing experience to a lively debate on this topic. They discuss the essence of BDD: is it the DSL, the automation frameworks, or is it more simply the conversations that take place as part of writing these scenarios/feature files? There are also limitations to BDD, so they discuss when it provides value, and when it might not. And they consider the question ‘just how expensive is it as an approach?'. Finally, they wrap up by considering whether BDD is simply misunderstood. Links from the podcast: Introducing BDD – Dan North The Tragedy of Given-When-Then – Chris Matts The Three Amigos The Value at the Intersection of TDD, DDD, and BDD – Darshan Satya

In this month's episode, we talk multi-cloud. This is a challenging topic, even the first step—nailing down what the term actually means—isn't easy. How does it differ from hybrid-cloud or poly-cloud? Does the term refer to the deployment approach for a single application or an entire organisation? We discuss the challenges of creating a multi-cloud architecture, dealing with multiple cloud vendors, and why you might follow this path. Colin is joined by colleagues James Heward and Robert Griffiths to compare thoughts and experiences gained from many years of architecting solutions for the cloud. Links from the podcast: Flexera 2022 State of the Cloud Report – Brian Adler Decoder: Polycloud – Thoughtworks White Paper: Thinking differently - the cloud as a value driver – Scott Logic

A relatively new architectural style for building web-based applications, micro-frontends are an extension of the popular microservices pattern where the vertical slice of functionality that a microservice provides is extended all the way to the front-end. With micro-frontends, you can more easily scale your development teams by composing applications from loosely coupled frontend components. In this month's episode, Colin is joined by Dean Kerr (Lead Developer, Scott Logic) and Sam Hogarth (Senior Software Engineer, Tesco Bank) to compare thoughts and experiences gained from many years of front-end development. Links from the podcast: Micro Frontends - Spotify Approach Iframes - Mostafa Biomee Microservice Websites - Gustaf Nilsson Kotte Webpack Module Federation Sites vs. Apps defined: the Documents-to-Applications Continuum - Aral Balkan You probably don't need a micro-frontend - Colin Eberhardt, Dean Kerr, Sam Hograth, Robat Williams, Chris Price, Chris Kurzeja

Microservices have become the standard architectural pattern for everything. It's a great and versatile pattern, but what people rarely talk about is the cost. You didn't think you got all that good stuff (polyglot, scalable, decoupled...) for free, did you? In this episode, we talk about the real-world impact of choosing this pattern and when it might not be the right choice.   In this month's episode, Colin is joined by Ian Sugar (Technical Principal, Scott Logic) and Harry Bellamy (Lead Developer, Scott Logic) to compare thoughts and experiences gained from years of working with microservices and their architectural antecedents. Links from the podcast: • Microservices for Life - Ian Sugar • Code reuse in microservices architecture - with Spring Boot - Bartosz Jedrzejewski • Designing Microservice architectures the right way - Michael Bryzek • When To Use Microservices (And When Not To!) - Sam Newman & Martin Fowler

Beyond the Hype is a brand new monthly podcast from the Scott Logic team, where we cast a practical eye over what is new and exciting in technology – everything from Kafka to Kubernetes, AI to APIs, microservices to micro-frontends. We look beyond the promises, the buzz and excitement to guide you towards the genuine value. We're an independent consultancy at Scott Logic, so there'll be no sales pitches or marketing speak – just impartial insights and a frank and open discussion. So, to get a clear-eyed view of the technology landscape, search for Beyond the Hype on Apple, Spotify or wherever you get your podcasts.