Amjad Masad and Haya Odeh, cofounders of Replit, join South Park Commons Partners Aditya Agarwal and Ruchi Sanghvi to share what it really took to build one of AI’s most important platforms, and how it’s now enabling a new wave of million-dollar founders.
Amjad breaks down why traditional ideas like ICP are breaking down in the age of AI, what’s actually changing under the hood of modern models, and why we may be reaching the limits of prompting. Haya shares the origin of “Seek Pain”—Replit’s most counterintuitive cultural principle—and how a relentless focus on what’s not working drives better products, faster learning, and stronger teams.
- Amjad Masad: https://www.linkedin.com/in/amjadmasad/
- Haya Odeh: https://www.linkedin.com/in/haya-odeh-b0725928/
- Ruchi Sanghvi: https://www.linkedin.com/in/rsanghvi/
- Aditya Agarwal: https://www.linkedin.com/in/adityaagarwal3/
- South Park Commons: https://www.linkedin.com/company/southparkcommons/
Apply to SPC: https://www.southparkcommons.com/apply
CHAPTERS:
(00:00:00) - Coming to America broke (and building anyway)
(00:04:27) - Early Replit proof points kept the mission alive
(00:07:02) - Cloud vs. local: why security tips the scales
(00:10:09) - Execute daily, predict quarterly
(00:11:35) - The 2023 roadmap Replit just finished executing
(00:16:01) - Agent 4 and the end of context amnesia
(00:22:01) - The death of the ICP
(00:24:47) - What actually changed in AI models December 2024
(00:28:52) - "Seek Pain"—Replit's most counterintuitive cultural value
(00:34:55) - Why consultants are the most mispriced AI-era hire
(00:38:00) - Co-founding with your partner—the honest answer
(00:43:25) - Make micro-predictions or get left behind by AI
(00:45:19) - Raising kids in a world you can't predict
