Catalog & Cocktails: The Honest, No-BS Data Podcast

Nachiket Mehta, VP of AI and Data Engineering who built ontologies and event-driven teams at Wayfair. He unpacks mapping happy vs unhappy paths, the $35M trailer story, shifting data teams left into domains, tiger teams and ops visits, and turning process maps into reusable ontologies for actionable enterprise AI.

Nachiket Mehta, an experienced data leader who built Wayfair’s enterprise insights work, explains why ontology, events, and culture matter for unlocking AI. He describes tying semantic layers to real-world fulfillment, embedding data teams into domains, and turning tribal knowledge into formal context. Short, practical, and focused on shifting data left for impact.

Vaibhav Gupta, co-creator of BAML and systems engineer who builds tooling and languages for reliable agentic and probabilistic code. He argues most agent code is ugly and fragile. He compares abstraction waves from web dev to LLMs. He explains why abstracting failure matters, demos typed prompts and schema enforcement, and shows how languages and tooling can make probabilistic systems robust.

A critique of how agentic AI is built today and why many implementations are fragile. Discussion of LLMs' probabilistic outputs creating tight feedback loops that amplify errors. Comparisons to gaming design for guardrails and feel. Proposal for new abstractions and typed patterns to give AI systems clearer structure and semantics.

Two colleagues rant over decision intelligence and how to codified decision lineage using context graphs. They debate MCP versus skills approaches and when interoperability matters. They argue companies have a work problem, not a data problem, and outline a three-stage maturity model for building a knowledge/context layer.

Juan Gorricho, a data and analytics leader with 25+ years at TD Bank, Visa and Disney, talks practical ways to turn data work into measurable business outcomes. He calls out vanity metrics, explains how to move from order-taking reports to decision partners, and advises building foundations by slice while earning trust with quick wins.

Juan Gorricho, a data and analytics executive with 25+ years leading strategy at companies like TD Bank and Disney, breaks down practical priorities. He urges counting real dollar impact over AI hype. Discussion hits trust and data literacy, cutting vanity metrics, focusing on a few critical data columns, and shifting reporting into decision-intelligence with product thinking.

Two friends unwind over beers while unpacking the biggest takeaways from Data Day Texas. They discuss practitioner-focused talks, the rise of agentic coding, and the librarian-style curiosity AI agents could adopt. Governance, context over models, and the push for interoperable semantics come up. Casual conference moments and debates about semantic layers and AI-ready data round out the conversation.

Dr. Irina Steenbeek, data practitioner and author focused on data lineage and aligning data and AI governance. She breaks down lineage at business, conceptual, logical and physical levels. Short practical takes on scoping lineage for value, when full technical lineage is unrealistic, and how lineage and governance share foundations like catalogs and metadata.

Practical realities of implementing data lineage, from when it helps and when it is overkill. How lineage untangles spaghetti systems to aid impact analysis and root-cause work. The overlap between data, lineage, and AI governance and why organizational factors often trump technical ones. Advice on focusing lineage efforts around use cases and necessary quality.