Sharing a new collaboration with the #SGLang community. https://t.co/Ef5HyfGZPF
Infer-forge uses Harness + Loop + Graph to keep 28-hour median task lifetimes fully traceable. Peak tasks in flight: 2→9. A full DeepSeek-V4-Pro serving project split into 38 pieces, each verified like a human engineer. And the harness catches kernel silent corruption on its own.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Why it matters
Long-horizon AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → runs usually fail on traceability and silent errors. Infer-forge reports 28-hour median task lifetimes kept traceable across 38 verified subtasks, with the agent harnessThe scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.Full definition → itself catching silent kernel corruption.
articleHarness Design for Long-Running Application DevelopmentAnthropic Engineering
articleHarness Engineering: Anatomy, Architecture, and Evolution of Coding Agents -- A Source-Code Study of Eleven SystemsPaul Barbaste, Tristan Darrigol, Germain Vu, Tom Wiltberger
articleBridging intent and execution in agentic systemswww.amazon.science