Brendan Foody
CEO and co-founder of Mercor, the expert labour marketplace used by AI labs for post-training data and evals. At the time of recording, the youngest unicorn founder. Grew Mercor from $1 to $400M ARR in 16 months — claimed fastest in history — before raising $100M at a $2B valuation. Started first business (Donut Dynasty — reselling Safeway doughnuts at school) in 8th grade; met his co-founders at age 14; started Mercor at 19 in January 2023, bootstrapping to $1M ARR before dropping out of college. Grew up in the Bay Area. Dyslexic; frames this as a strength for differentiated market thinking. Based in San Francisco.
Core positions
Argues the scarce input for AI progress is no longer raw text but rubrics — structured, reusable ways of measuring success — and that AI labs should benchmark against economically valuable, expert-derived tasks (Mercor’s APEX index) rather than academic benchmarks like GPQA or IMO. Expects a large share of knowledge work to convert into one-time ‘teach the model’ investments, mirroring software’s shift from recurring bespoke labour to a fixed cost reused at scale, and predicts a new job category — training agents and building RL environments — will employ a majority of high-end knowledge workers within five years. Frames job survival under AI as a question of price elasticity of demand rather than raw automatability.
Episodes
- Brendan Foody on Evals, the Expert Labour Market, and Mercor's Rise — Lenny’s Podcast
- Brendan Foody on Teaching AI and the Future of Knowledge Work — Conversations with Tyler