Data & AI Mastery

👉 Discover how Cambridge Spark helps organisations build the data and AI capabilities needed to turn strategy into measurable impact: cambridgespark.com On this week's episode of Data & AI Mastery, Dr Raoul-Gabriel Urma is joined by Sarah Self, AI Director at Aviva. Sarah brings a rare dual perspective to the conversation of data and AI, having spent nearly a decade in cybersecurity before moving into AI leadership. In this episode, she shares how those two disciplines share the same foundations: people, process and technology, and why that framework is central to delivering responsible, scalable AI. Sarah discusses how Aviva built its own internal AI platform to embed guardrails from day one and why she views risk management and customer value not as a trade-off but as a single, unified objective. She also makes a compelling case for democratising AI across the workforce, using the term "organ rejection" to describe what happens when people are left behind in a transformation. The episode closes with two standout production use cases: a claims summarisation solution that cut customer hold times by over 50% and a medical underwriting tool delivering 99% accuracy, both of which drive efficiency and customer satisfaction simultaneously. Follow Data and AI Mastery on Apple Podcasts, Spotify or YouTube so you never miss an episode. Chapter Markers (00:00) - Opening Insight: Experimentation With Purpose (02:44) - The Parallels Between Cybersecurity and AI (05:46) - Balancing Risk and Customer Outcomes at Aviva (10:29) - Democratising AI: Why Workforce Inclusion Matters (14:28) - Identifying High-Value AI Use Cases (17:27) - Balancing Experimentation With Production: The Portfolio Approach (20:16) - Remaining Challenges and Keeping Perspective on the Hype (23:48) - Quickfire Round: Sarah's AI Aha Moment (25:21) - Raoul’s Reflections Useful Links Connect with Sarah Self on LinkedIn: https://www.linkedin.com/in/sarah-self-ba0b0642/ Follow Raoul for more AI insights on LinkedIn: https://www.linkedin.com/in/raoulurma/ Explore Cambridge Spark’s AI upskilling programmes at https://www.cambridgespark.com

👉 Discover how Cambridge Spark helps organisations build the data and AI capabilities needed to turn strategy into measurable impact: cambridgespark.com On this week's episode of Data & AI Mastery, Mark Bramwell, Chief Digital and Information Officer at Oxford Said Business School, joins Dr Raoul-Gabriel Urma to share how one of the world's oldest universities is navigating one of its biggest challenges: deploying AI across a complex, multi-persona institution while simultaneously preparing students for an AI-driven workforce. Mark unpacks Oxford's deliberate portfolio approach to AI tools, why they moved from per-user licensing to consumption-based models, and how they are measuring real business outcomes rather than treating AI as something categorically different from the rest of the tech stack. From a faculty-led AI digital avatar teaching courses in any language, 24/7, to an AI assistant enhancing the MBA student experience, this conversation is grounded in genuine implementation rather than theory. If you lead a complex organisation grappling with AI adoption, governance, and change management at scale, this episode is essential listening. Follow Data and AI Mastery on Apple Podcasts, Spotify or YouTube so you never miss an episode. Chapter Markers (00:00) - Opening Insight: Treating AI Like Any Other Tool (04:02) - Driving Innovation in a Centuries-Old Institution (05:24) - Oxford's AI Strategy: The Portfolio Approach (09:52) - Risks and Lessons Learned from the Consumption Model (12:03) - Preparing Students for an AI-Driven Workforce (17:44) - Governing AI Adoption at Scale (22:58) - Change Adoption and Getting Everyone on the Journey (25:51) - The Future of AI in Higher Education (27:44) - Rapid-Fire Personal Questions (30:06) - Raoul's Key Takeaways Useful Links Connect with Mark Bramwell on LinkedIn: https://www.linkedin.com/in/mark-bramwell-fbcs-308b4913/ Follow Raoul for more AI insights on LinkedIn: https://www.linkedin.com/in/raoulurma/ Explore Cambridge Spark’s AI upskilling programmes at https://www.cambridgespark.com

👉 Discover how Cambridge Spark helps organisations build the data and AI capabilities needed to turn strategy into measurable impact: cambridgespark.com On this week's episode of Data & AI Mastery, Miryem Salah, Director of Digital, Data and Transformation at VodafoneThree, joins Raoul Gabriel-Urma to share the story behind one of the most ambitious AI strategies in UK telecoms. Miryem breaks down how her team approached building a forward-looking AI strategy from first principles, why most organisations fall into the FOMO trap, and how VodafoneThree is thinking about the balance between humans and AI in a post-merger business. She also shares her thinking on governed agility, the importance of data literacy at every level of leadership, and why getting foundations right matters more than chasing the next shiny tool. The conversation also touches on diversity in data and tech, Miryem's Women in Data network spanning 100,000 employees across VodafoneThree; and her genuine excitement about the rise of physical AI. A frank, experience-led conversation for senior data and AI leaders navigating large-scale transformation. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers (00:00) - Introduction and Miryem's role at VodafoneThree (02:00) - Inside the Vodafone and Three UK merger (06:00) - Misconceptions about data teams and data ownership (10:00) - Operating model, human versus AI, and the sell-build-run framework (15:00) - Measuring success in early-stage AI deployment (19:00) - Risk management and when to put AI in front of customers (24:00) - Quickfire round: diversity, STEM, music and the magic wand question (27:00) - Raoul on AI and the future of education (28:30) - Key takeaways Useful Links Connect with Miryem Salah on LinkedIn: https://uk.linkedin.com/in/miryem-salah-7a3ba633 Follow Raoul for more AI insights on LinkedIn: https://www.linkedin.com/in/raoulurma/ Explore Cambridge Spark’s AI upskilling programmes at https://www.cambridgespark.com

👉 Discover how Cambridge Spark helps organisations build the data and AI capabilities needed to deliver responsible, real-world impact: cambridgespark.com In this episode of Data & AI Mastery, host Dr. Raoul-Gabriel Urma is joined by Edmund Towers, Head of Advanced Analytics & Data Science Units at the Financial Conduct Authority (FCA), to explore how data science and AI are being used to protect consumers, fight financial crime, and enable safe innovation across financial services. Ed leads advanced analytics and data science across the FCA, working at the intersection of technology, regulation, and public good. From identifying scams and money laundering to shaping the UK’s regulatory approach to AI, his work demonstrates how responsible use of data and AI can deliver real societal impact at scale. Together, Raoul and Edmund unpack how regulators can both enable innovation and protect consumers, why outcomes-based regulation matters, and how initiatives like AI Live Testing are helping firms move from proof-of-concept to safe deployment. This episode is essential listening for data and AI leaders, compliance and risk professionals, policymakers, and executives who want to understand how AI can be used responsibly to deliver trust, safety, and innovation, not just efficiency. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers (01:40) — Edmund Towers’ role at the FCA (04:20) — Why the FCA must both regulate and innovate (08:40) — Introducing the FCA’s AI Live Testing initiative (12:40) — FCA strategic priorities: financial crime, consumers, innovation (16:00) — Synthetic data and collaboration with the Alan Turing Institute (19:20) — Reducing friction in complex financial decisions (22:30) — Quick-fire round: favourite language, subjects, and music (23:30) — Raoul’s reflections on regulation, innovation, and collaboration Useful Links Connect with Edmund on LinkedIn Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com

Discover how Cambridge Spark helps organisations build the data and AI capabilities needed to deliver measurable business impact: cambridgespark.com In this episode of Data & AI Mastery, host Dr. Raoul-Gabriel Urma is joined by Conny Ploth, VP of Global AI Transformation in the financial services sector, to unpack what it really takes to deliver measurable, people-first AI transformation in complex, regulated organisations. Conny brings a unique perspective shaped by a career spanning microfinance, consulting, ESG, and large-scale financial institutions. Her message is clear: if you can’t measure it, don’t start it and AI transformation only succeeds when data, people, and processes move together. Together, they explore how leaders can move beyond AI hype and focus on impact, adoption, and sustainable value creation. Listeners will learn why measurement and baselines are non-negotiable before starting any AI initiative, how to balance quick wins (task automation) with longer-term transformational bets and how legacy organisations can compete with AI-native startups by creating safe spaces for experimentation. This episode is packed with practical guidance for CIOs, transformation leaders, and executives who want AI to deliver real outcomes — not just prototypes. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers: (01:40) — Conny Ploth’s career journey into AI leadership (06:00) — The three AI value levers: efficiency, growth, experience (10:50) — Governance, guardrails, and balancing access with protection (14:30) — Setting the AI vision and aligning with business strategy (18:40) — Identifying quick wins vs long-term strategic initiatives (24:00) — Running AI as a portfolio of initiatives (28:00) — Upskilling at scale: why experience beats traditional training (33:50) — Quick-fire round: fitness, routines, and personal habits (35:40) — Raoul’s closing reflections: people, process, and purpose Useful Links: Connect with Conny on LinkedIn Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com

Discover how Cambridge Spark helps leaders and teams build the data and AI skills needed to operate in high-stakes, real-world environments: cambridgespark.com In this episode of Data & AI Mastery, host Dr. Raoul-Gabriel Urma is joined by Andrew Feustel, former NASA astronaut, ISS Commander, and veteran of three space missions. Andrew has spent over two decades at NASA, flying on the Space Shuttle, leading missions aboard the International Space Station, and performing complex spacewalks to repair critical infrastructure like the Hubble Space Telescope. In this conversation, he shares a rare, human perspective on decision-making in the most extreme data environments imaginable: outer space. Together, they explore what space exploration can teach business and technology leaders about trust, automation, AI, and the limits of data-driven systems. Listeners will learn how astronauts rely on data, telemetry, and human judgment during high-risk missions, why data never perfectly matches reality, and why that matters in critical environments and why humans must remain in the loop when decisions carry irreversible consequences This episode offers powerful lessons for leaders navigating AI adoption, risk management, and decision-making, reminding us that technology is only as effective as the humans who design, trust, and use it. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers: (04:40) — Why data is central to every space mission (09:40) — Trust, automation, and critical systems in spaceflight (11:00) — Why we don’t fully trust autonomous systems (yet) (14:20) — Why AI is essential for space telescopes and discovery (16:40) — What makes NASA exceptional at preparing people (19:10) — How AI is already accelerating space innovation (20:30) — Automated rocket landings and decision systems (23:50) — Quick-fire round: unforgettable space memories (28:00) — Raoul’s reflections: trust, AI, and human-in-the-loop systems Useful Links: Connect with Andrew on LinkedIn Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com

Learn how Cambridge Spark helps organisations and leaders develop the data and AI skills that drive real transformation: cambridgespark.com Ever wondered how Formula 1 teams make lightning-fast decisions at 200mph? In this episode, host Raoul-Gabriel Urma sits down with Ruth Buscombe, former race strategist for Scuderia Ferrari, Sauber, and Haas F1, to unpack the incredible world of real-time data strategy in one of the most high-pressure environments on Earth. Ruth shares what it's really like making million-dollar calls in milliseconds, how her team processes 1.1 million data points per second from each car, and why the best lessons come from spectacular failures (not champagne-soaked victories). One of the most fascinating parts of the discussion is when Ruth explains the eternal tension between data and gut feel, and why human intuition still matters even when you have the best models in the world. Spoiler: your "gut" is really just your brain's LLM of past experience. They also explore how F1 teams decide which sensors and data points actually move the needle when margins come down to 0.012 seconds, and what it's like to make high-stakes pit-stop calls with incomplete information (sound familiar, business leaders?). Ruth also shares some powerful stories about ‘failing forward’, including why Mercedes boss Toto Wolff says "the day we fail is the day our competitors should fear us most." Plus, they look at how AI is transforming sports broadcasting, with F1 now able to search 75 years of race footage using simple natural-language queries. And of course, they wrap up with a quick-fire round covering everything from Python vs MATLAB to favourite subjects at Cambridge and yes, Taylor Swift workout playlists. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. You can check out Formula 1's official site to learn more about the sport, or head to F1 TV if you want to dive deeper into the data and analysis Ruth works on. Whether you're leading a startup, managing a data team, or just fascinated by how top performers think under pressure, this conversation is packed with insights you can actually use. Chapters 00:00 - Introduction: Learning from F1's best failures 02:00 - Ruth's journey from Cambridge engineering to the F1 pit wall 03:00 - The insane scale of F1 data (1.1M points/second!) 06:00 - How do you decide what data to collect? 09:00 - Three pillars: Performance, competitor analysis & race strategy 13:00 - When should you trust data vs. gut instinct? 17:00 - Managing risk like a portfolio manager 20:00 - Creating a culture that embraces pressure 23:00 - AI transforming F1 broadcasting & fan experience 26:00 - Quick-fire: Python, physics & being a Swiftie 27:00 - Key takeaways for business leaders Links Connect with Ruth on LinkedIn Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com Visit Formula 1's official site at formula1.com Visit F1 TV at f1tv.formula1.com

Learn more about how CambridgeSpark.com is helping organisations and professionals master Data & AI skills to stay ahead in the digital era. In this episode of Data & AI Mastery, host Dr. Raoul-Gabriel Urma sits down with Dr. Petar Veličković, Senior Staff Research Scientist at Google DeepMind and Affiliate Lecturer at the University of Cambridge. Together they explore how cutting-edge AI research transitions from theory to practice, spanning from breakthroughs in graph neural networks that power Google Maps worldwide, to the use of AI as a discovery partner in mathematics. Petar shares his journey from Serbia to Cambridge to DeepMind, and the pivotal lessons learned at every stage. Listeners will also hear how DeepMind’s graph machine learning models improve global travel-time predictions in Google Maps, how AI can accelerate scientific discovery, including mathematical proofs and what the concept of an “AI scientist” might mean for the future of research. Whether you’re a data scientist, ML engineer, or executive exploring AI strategy, this episode offers a masterclass in applying AI innovation with purpose and impact. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers: (06:00) - Inside DeepMind; what a Senior Staff Research Scientist does (13:00) - From research to real-world impact: challenges & latency trade-offs (19:00) - AI for mathematics; collaborating with top researchers (25:00) - Defining the “AI scientist” and the future of autonomous research (30:00) - What’s next for AI agents and self-directed systems (31:00) - Quick-fire round; programming languages, school favourites & music (33:00) - Raoul’s takeaways; user focus, AI as a thought partner, safe deployment Useful Links: Connect with Petar on LinkedIn Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com

Visit cambridgespark.com to explore how we help organisations upskill their workforce in Data & AI. In this episode of Data & AI Mastery, Dr. Raoul-Gabriel Urma speaks with Aidan Innes, Head of Data Standards at Nuffield Health, the UK’s largest healthcare charity. Aidan shares his unique journey from sports science into data leadership, revealing how he built a strategy that connects patient outcomes, data literacy, and advanced analytics to drive both social and financial impact. Elsewhere in the episode the pair discuss building outcome frameworks that scale across a complex healthcare organisation and how Nuffield Health is leveraging data apprenticeships to boost literacy, save millions, and empower staff. Whether you’re a healthcare professional, a data leader, or simply someone passionate about how data drives impact, this episode is packed with insights and actionable takeaways. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers: (05:00) – COVID-19 rehabilitation programme and measuring outcomes (08:00) – Scaling outcome frameworks across the organisation (12:00) – Technology, governance, and people in data strategy (16:00) – Measuring ROI and success stories in procurement (18:00) – Advanced analytics and award-winning projects (22:00) – Quick-fire round: contrarian views, learning habits, and personal insights (24:00) – Aidan’s vision for the future of data in healthcare (26:00) – Closing reflections and takeaways Useful Links: Connect with Aidan on LinkedIn Visit the Nuffield Health Website Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programmes at cambridgespark.com

Start your data & AI transformation journey with Cambridge Spark. In this episode of Data & AI Mastery, Dr. Raoul-Gabriel Urma speaks with Matthew Cloke, Chief Technology Officer at Endava, about how one of the world’s leading technology companies is embracing AI at scale. Matt shares how Endava created an AI committee to set governance and strategy, built cross-functional champions networks and celebrated AI heroes to drive adoption, and invested in ChatGPT Enterprise across 8,000 employees, achieving 85–90% daily usage. The pair also dive into the business case for AI and how to secure executive buy-in and why creating "wake-up moments" for executives is critical. Whether you’re a senior leader shaping AI strategy or a professional curious about scaling adoption, this conversation is packed with actionable insights. Be sure to follow Data & AI Mastery wherever you listen to your podcasts to never miss an episode. Chapter Markers: (03:45) Who owns the AI agenda? Inside Endava’s AI committee (10:30) The Champions Network & AI Heroes (15:20) Making the business case for ChatGPT Enterprise (25:30) “Wake up, crawl, walk, run”: driving executive aha moments (29:30) Balancing wait-and-see vs. first-mover advantage (32:40) Quickfire round: contrarian views, environment, favourite language & more (38:40) Closing reflections & key takeaways Useful Links: Connect with Matthew on LinkedIn Visit the Endava Website Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programs at cambridgespark.com

Learn how Cambridge Spark can help your business transform with data and AI: https://cambridgespark.com In this episode of Data & AI Mastery, host Dr. Raoul-Gabriel Urma sits down with Daniel Hulme, Chief AI Officer at WPP, to explore a bold rethinking of artificial intelligence—beyond automation and into adaptive, evolving systems. Daniel shares how marketing is being reengineered through AI—dynamic segmentation, synthetic audiences, brand-safe content generation, and more—and how businesses must focus on frictions rather than blindly chasing trends. Drawing on decades of experience, he also reveals a powerful framework for AI defensibility built on data, talent, and leadership. This is a must-listen for executives, AI practitioners, and innovators eager to move past hype and into action. 🎧 Enjoyed this episode? Subscribe and follow to catch more discussions with data & AI leaders who are transforming their industries. Chapter Markers: (05:50) The case for goal-directed adaptive systems (08:05) Why businesses chase tech hype instead of solving problems (12:34) Case study: Using AI to augment creativity at scale (15:07) Micro-moments & the future of personalised, timely marketing (17:00) Real-world transformation challenges (and how to overcome them) (19:05) What gives AI efforts defensibility in business (21:39) Biggest misconceptions business leaders have about AI (24:46) Daniel on staying ahead: predictions, neuromorphic AI, philosophy Useful Links: Connect with Daniel Hulme on LinkedIn Visit the WPP Website Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programs at cambridgespark.com

Transform your workforce with cutting-edge data & AI skills — visit cambridgespark.com to get started today. In this episode of Data & AI Mastery, Dr. Raoul-Gabriel Urma sits down with Jeff Boudier, who leads Product and Growth at Hugging Face, the world’s leading open platform for AI builders. Throughout the episode, the pair explore how enterprises are turning to smaller, specialised open models to solve real business problems; the importance of maintaining control over your data; and how Hugging Face is helping enterprises bring AI safely into production. Whether you’re a CTO navigating how to deploy AI at scale or an engineer eager to sharpen your edge, this episode delivers powerful takeaways on using data & AI strategically — not just for experimentation, but for driving real ROI. Subscribe to Data & AI Mastery and never miss insights from global leaders pushing the boundaries of what’s possible with data. Chapter Markers: (05:50) Real-world ROI: Financial services & patent tech using smaller open models (08:50) When to adopt specialised vs. large models (10:50) The AI adoption curve & why general models aren’t one-size-fits-all (12:30) Protecting your data: governance, builder culture, and avoiding lock-in (14:30) How Hugging Face keeps up with the latest developments in AI (17:50) Supporting enterprises: from rogue IT to secure, compliant AI (21:30) Security, safe tensors & scanning models for vulnerabilities (24:40) Quick-fire round: Jeff’s favourite programming language, subjects & music genre Useful Links: Connect with Jeff Boudier on LinkedIn Visit the Hugging Face Website Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programs at cambridgespark.com

Want to equip your workforce with the data and AI skills they need to drive real transformation? Cambridge Spark partners with global organisations to build high-impact, scalable training programmes that deliver measurable business outcomes. Learn more at cambridgespark.com. In this episode of Data & AI Mastery, Dr. Raoul-Gabriel Urma speaks with Roxane Heaton, Managing Director of Digital Colleague Experience at Barclays, about what it means to design digital systems for real people—at scale. Roxane shares her unique journey across the public and private sectors, from the Cabinet Office to Macmillan and now at Barclays, and brings an unshakable focus on user-centered innovation. In a world obsessed with digital transformation, Roxane challenges conventional thinking by arguing that AI doesn’t replace thinking—it deepens it. Whether you're a digital leader, HR strategist, or AI practitioner, Roxane offers an inspiring and grounded vision for inclusive, impactful transformation. Tune in to discover how to unlock human potential through data-driven design. Chapter Markers: (05:46) How to Measure the Impact of Digital Tools (08:32) Using Passive vs Active Data in AI Design (10:47) Automation, Copilots & Rethinking Workflows (13:07) Why Upskilling Requires More Than Just Tools (16:33) The Role of Ethics and Data Governance (19:51) Storytelling as a Leadership Skill in the AI Era (20:21) Why Design Thinking is the Must-Have Mindset (26:01) Raoul’s Reflections Useful Links: Connect with Roxane Heaton on LinkedIn Visit the Barclays Website Follow Raoul for more AI insights on LinkedIn Explore Cambridge Spark’s AI upskilling programs at cambridgespark.com