20VC: Mercor CPO on Revenue Concentration from Frontier Labs | Why Large Enterprise is Scared to Partner with Frontier Labs | Why Small Specialised Models is the Future with Osvald Nitski
Summary
Harry Stebbings interviews Oswald Nitsky, CPO at McCaw, about AI data strategies and product management. They discuss whether open-source models cannibalize frontier data needs, the shift toward specialized enterprise models, and the evolving role of PMs in an AI-driven world. Nitsky also shares insights on hiring, managing hyper-growth, and the rising demand for RL environment data.
Topics discussed
Sponsors: Asana, Zero Hash, and Framer
Does open source cannibalize proprietary data providers?
Enterprise AI adoption and specialized models
AI ROI, token spend, and budget optimization
Product management in the age of AI agents
Determining enduring demand and data team sophistication
Hiring strategies: Seniority, agency, and the 'Mercor Mafia'
Interview processes and the risk of AI-assisted creativity
Mercor's product structure: Marketplace vs. Managed Services
Expert experience, retention, and margin dynamics
Democratizing data and the rise of RL environment data
Price sensitivity and the unbundled data provider landscape
Competition, big tech distractions, and cybersecurity data
Hiring in SF, company culture, and quickfire advice
Mercor's bull case and the future of robotics
Closing remarks and sponsor ads
Listen ad-free on Castria