Thrive Researcher Shares Talks on Frontier Context Engineering and Model Interpretability
thesephist · x · 2026-08-11
Linus Lee (@thesephist), a researcher at Thrive Capital, recently uploaded two of his technical talks from earlier this year to his blog.
The talks primarily cover two areas:
- Notations & Interp: Musings on the magic of notations, model interpretability, and some old side projects rarely discussed in public.
- Context Engineering: Shares context engineering and retrieval-specific harness work that was found effective at Thrive earlier this year.
More from coding & agent
- AI Agents Autonomously Shipping Code to Prod Feels Inevitable — nbaschez · 2026-08-11
- Docker Is Isolation, Not Enforcement: Why Containers Aren't Sandboxes — max_paperclips · 2026-08-11
- Awesome A2A: A Curated GitHub List for the Agent2Agent Protocol Ecosystem — tom_doerr · 2026-08-11
- Stop Pre-injecting Context: Let Agents Search to Reduce Errors — siddharthnibjiya · 2026-08-11
- Orchestrating 12 Claude Agents to Write a Paper and Build a Website — sebkrier · 2026-08-11
- Opus 5 + Higgsfield One-Shot Generates 3D Parkour Game with Zero Hand-Modeled Assets — Vjeux · 2026-08-11