Sebastian Raschka: the harness is part of training, LLMs can't absorb it
rasbt · x · 2026-11-21
Responding to the claim that LLMs will eventually absorb agent harnesses and make them obsolete, Sebastian Raschka pushes back with a technical argument: LLMs are specifically fine-tuned to work in particular harnesses with particular tools—the harness is part of the training setup, so the model can't absorb it. His analogy: it's like expecting the CPU/GPU to absorb the operating system. The quoted post by llmwizard first mocks the idea as not remotely "common" enough to worry about.
More from coding & agent
- MCP Events Arrive: Proactive Agents via a Single Natural-Language Prompt — DavidWells · 2026-11-21
- Dev clones Claude Meetings using Kimi K3.1 Pro on Modal hours after Anthropic's launch — AAAzzam · 2026-11-21
- Astra 6.2 generates an ocean simulation with buoyant balls in one HTML file — mark_k · 2026-11-21
- LangChain: scheduled runs hit 84% of agent runs after five weeks deployed — LangChain · 2026-11-21
- Dev insight: the most important part of agent harnesses is stopping AI shortcuts — clemnt · 2026-11-21
- Jeremy Howard: tying Neon to Antigravity will limit its success — jeremyphoward · 2026-11-21