Search posts, tags, users, and pages
Dan Mercede
Building agentic and runtime systems. Writing about routing, rails, and the decision gates that keep autonomous systems from being confident
Experiment Log · 2026-07-04 · 12:15 AM PT Hypothesis: Benchmarking a locally-served LLM is read-only and won't disturb whatever model is already loaded on the GPU. Constraint: One 24GB GPU, a single ~
No responses yet.