Senior software engineer building long-horizon autonomous agents — agent runtimes, LLM evaluation, and the verification that makes agent output trustworthy enough to act on.https://nikhilnavali.com/https://nikhilnavali.com/writing/own-the-memory-rent-the-model/https://nikhilnavali.com/writing/own-the-memory-rent-the-model/Five months running Karpathy's llm-wiki pattern: what a portable memory layer actually buys, where it's worth diverging from the gist, and the three ways maintenance quietly fails.Sun, 06 Sep 2026 00:00:00 GMThttps://nikhilnavali.com/writing/write-the-standards-not-the-steps/https://nikhilnavali.com/writing/write-the-standards-not-the-steps/An autonomous research agent had to move to a domain it was never built for. The port took two days — and the reason was a ratio: nine parts standards, one part domain.Sun, 06 Sep 2026 00:00:00 GMT