[Online] Why Your Agents Need Data Contracts Too

Continuing our DPM webinar series aimed at upskilling & knowledge-sharing, we're joined by Andriy Mandyev - Data/AI Architect and Strategy Consultant! Every team is building agents. But agents are only as reliable as the data they consume — and most teams haven't designed their data with agents in mind. Data contracts have become standard practice for data pipelines. The argument is simple: producers and consumers should agree on schema, semantics, and SLAs. But as AI agents increasingly depend on structured data feeds, the same logic applies — with higher stakes. An agent that hits an unexpected null doesn't just fail a dashboard. It may take a wrong action. Andriy Mandyev has spent 15+ years building data organisations and architectures at scale — across finance (Société Générale), retail (Decathlon), and start-ups (Bryj) — and has watched this problem emerge from the ground up. What we'll unpack Why data contracts matter for agentic AI, not just data pipelinesWhat breaks first when agents encounter inconsistent or undocumented dataHow to design data interfaces that agents can depend onWho owns the contract between data producers and agent buildersLessons from building at scale across Société Générale, Decathlon, and Bryj What you'll walk away with A framework for thinking about data contracts in an agentic contextPractical patterns for making your data agent-readyAn understanding of where the contract sits in your architectureQuestions to ask your team before your next agent build Perfect for: Data product managers, data architects, platform engineers, and AI/ML teams building or evaluating agentic systems. If you've ever had an agent silently fail because upstream data changed, this one's for you. Whether you're just starting to think about agents or already running them in production, this session will give you a practical lens for the data layer that most teams miss. About our speaker Andriy Mandyev brings over 15 years of experience in the tech industry, spanning the evolution of software from high-performance distributed systems to data platforms, analytics, machine learning, and agentic AI. Having built and led data organisations at scale, designed enterprise data architectures, and delivered data services across hundreds of operational and analytical use cases, Andriy's career has moved from development to leadership across diverse sectors including Finance (Société Générale), start-ups (Bryj), and Retail (Decathlon). Regardless of the role, the lens remains the same: the art of scaling — be it systems, teams, or products. How to join This livestream will be held privately inside our new digital space! Make sure you've joined us on Circle - signing up takes 30 seconds and is free. About the Data & AI Product community Data and AI product management is still a young discipline, and there aren't many spaces dedicated to learning from peers. So we started this meetup in 2023 to change that! Since then, it's grown into a vibrant community with chapters in London, Barcelona, and Paris. Whether or not "product" is in your job title, if you're involved in shaping data, analytics, and AI initiatives (e.g. product managers, strategists, BAs, data scientists, engineers, analysts) you'll find like-minded people here. This is an informal, welcoming space to swap lessons, share challenges, and enjoy drinks and snacks along the way 😊 You can meet us in person at our in-person meetups, but also join our global community where we have expert webinars, training courses, job postings, resources, and discussions. Quick FAQ Will this event be recorded? Yes, we will upload the recording here, along with previous DPM Community webinarsIf I sign up to this event, will I be opting into any marketing emails? No. We only use the Luma mailing list to invite you to upcoming community events and share community updates (like our end of year membership survey). You can always opt out of the Luma calendar.Can I mention this to a friend or colleague? Yes, of course! If they're based in London, send them this event. If they're not, it might be better to send them the LinkedIn Live link directly, so they don't get invited to all our in-person meetups too.Why are you doing this? Data & AI Product Management is a nascent field, and we created this community to bring together practitioners and help share our skills and experiences with those newer to DPM.