How GridCore shares one 16 GB GPU between chat, agents and background jobs: design rules, VRAM estimates and the bugs only real hardware showed.
A GPU Scheduler for Local LLMs: What Broke on Real Hardware Nick Blokhin Software Architect & Lead AI Engineer, 20+ years in production. Techstars Boston '15 · 12+ shipped apps · 32,000+ on YouTube · My weekly AI newsletter Home Projects AI Notes About Contact / Notes / A GPU Scheduler for Local LLMs: What Broke on Real Hardware A GPU Scheduler for Local LLMs: What Broke on Real Hardware 2026.10.01 Several things want my GPU at the same time: the chat I'm typing into, a coding agent, an indexer working through incoming documents, and a model that reads my screenshots. There is one card — a
How GridCore shares one 16 GB GPU between chat, agents and background jobs: design rules, VRAM estimates and the bugs only real hardware showed.
A GPU Scheduler for Local LLMs: What Broke on Real Hardware Nick Blokhin Software Architect & Lead AI Engineer, 20+ years in production. Techstars Boston '15 · 12+ shipped apps · 32,000+ on YouTube · My weekly AI newsletter Home Projects AI Notes About Contact / Notes / A GPU Scheduler for Local LLMs: What Broke on Real Hardware A GPU Scheduler for Local LLMs: What Broke on Real Hardware 2026.10.01 Several things want my GPU at the same time: the chat I'm typing into, a coding agent, an indexer working through incoming documents, and a model that reads my screenshots. There is one card — a
Related listings
Drop in a redacted PDF and see if the text under the black boxes can still be copied out. Free, runs entirely in your br...
