Press Space to continue
Press Space to continue
For a lot of Together AI's products, agents are already the majority user. Sit with that before the implication. The primary consumer of the product is not a person reading a page. It is a model reading a page, and Charles Zedlewski is building for that reader first. Which changes what quality assurance even means. Human UX gets validated through research sessions, interviews, watching someone struggle. Slow, expensive, and periodic. You learn your onboarding is broken weeks after shipping…
Aakash Gupta@aakashgupta4d ago
Together AI just raised $800M at an $8.3B valuation. I got their product team to open up the actual repo they run on: 0:00 - Slop is the new party foul 1:44 - Why individual output backfired 3:26 - Inside the shared product repo 8:57 - Who owns the context files 10:41 - Team skills vs personal skills 11:57 - Picking a harness and a model 15:17 - Feature research running live 16:57 - What stays human in the loop 19:56 - The PRD skill that interviews you 23:09 - Half a day of research in 5 min 24:25 - Their customer insights MCP 27:31 - What a good PRD looks like now 32:02 - Reading the finished PRD 35:07 - Every repo in one place 40:24 - Context is a hierarchy, not a pool 41:20 - Build your own orchestrator 44:19 - Agent evals on shipped features 49:41 - Design validated every few hours 52:26 - Where PM ends, engineer begins 55:10 - What it actually cost them
View
