@turingbook: The greatest potential value of Harness (Agent) may be going into the real work scenarios of seasoned experts across industries, observing how they work and think, and recording and "distilling" their valuable experience—data that was previously largely tacit.
Summary
The tweet highlights the potential of agent harnesses to capture and distill experts' tacit knowledge as new training data, referencing a Berkeley AI Summit talk by Jianfeng Gao on agentic modeling as an emerging AI paradigm.
View Cached Full Text
Cached at: 08/03/26, 07:39 AM
The greatest value of Harness (Agent) may be going into the actual working scenarios of seasoned experts across industries, observing how they work and think, and recording and “distilling” their valuable experience—data that was previously tacit.
空谷 Arvin Xu (@arvin17x): The biggest takeaway from today’s Berkeley AI Summit 2026 was Professor Jianfeng Gao’s talk.
A new paradigm for AI Modeling has emerged:
- New training data: from harness to agentic modeling — playing the same role that internet data did for pretrained models back then
- New Scaling Law: the AI industry is now producing massive amounts of AI Agent Harness at a furious pace every day
Similar Articles
@Potatoloogs: https://x.com/Potatoloogs/status/2057391224592667051
This article deeply analyzes the concept of Agent Harness, which is the engineering infrastructure wrapped around an LLM, including 12 components such as orchestration loops, tool calling, memory systems, context management, etc. The article cites practices from companies like Anthropic, OpenAI, and LangChain, arguing for the critical role of the harness in production-grade AI agents.
@xiaogaifun: The most thorough talk about Harness. This is probably the most thorough sharing I've seen about Harness Engineering, I recommend everyone watch it. Video link: https://podwise.ai/dashboard/episodes/8013289…
This article deeply explains the concept of Harness Engineering through a talk by IBM engineer Tejas Kumar, which involves adding deterministic infrastructure (such as tool registries, context management, guardrails, and validation loops) to AI Agents to solve model out-of-control and hallucination problems, ensuring stable task execution.
@rohanpaul_ai: So much recent work and research papers points to the same thing: the "harness" is becoming the real capability layer. …
Offloop's multi-agent harness achieves state-of-the-art results on GDPval, GDP.pdf, and JobBench benchmarks, outperforming Claude Code and Codex at 1/3 to 1/10 the cost per task, targeting $2.4T in US knowledge work.
@dotey: Building an Agent Harness itself is no longer valuable—no matter how hard you try, you can't compete with model companies. Once the model upgrades, much of your work becomes obsolete. But building solutions on top of a mature Agent Harness has great potential. MCP only solves the connectivity problem, Skills only solves the domain knowledge problem…
The author argues that directly developing an Agent Harness is of little value because model companies will dominate, but building applications in vertical domains on top of mature frameworks still offers significant opportunities. It requires redesigning AI-native workflows, UI/UX, and data organization.
@arvin17x: The biggest takeaway today at Berkeley AI Summit 2026 came from Professor Jianfeng Gao's talk. A new paradigm for AI Modeling has emerged: - New training data: from harness to agentic modeli…
The author summarized Professor Jianfeng Gao's presentation at the Berkeley AI Summit 2026, pointing out that the new AI modeling paradigm is based on Agentic Modeling and a new data flywheel, and emphasizing that Harness will become the key to application moats.