@jakevin7: Very happy to have witnessed and participated in this process. “Everything in human civilization can be redone (obviously we have observed that all SaaS can be redone, and we will soon see more, and it's not just computer software)”
Summary
A comment about AI: the author marvels that everything in human civilization can be redone with AI, and discusses the roles of the harness, system prompt, and toolset in model training.
View Cached Full Text
Cached at: 08/11/26, 05:45 AM
I’m thrilled to personally witness & participate in this process.
“Everything in human civilization can be redone (clearly we’ve already observed that all SaaS can be redone, and soon we’ll see much more — and certainly not just computer software)”
stdrc (@istrdc): You might not understand what a harness actually is
If it’s just that bit of system prompt and tools, then obviously models can learn them, and from then on you only need a tool list + tool implementation. Furthermore, once models get a bit more powerful, we’ll only need a bash tool. I’ve participated in Kimi K2.5 model training, and started writing from an empty system prompt, toolset, and loop.
Similar Articles
@ba_niu80557: https://x.com/ba_niu80557/status/2073413449930207662
Superpowers 6 open-source project shows that AI can not only write code but also autonomously optimize development workflows (such as auditing, merging tasks, reducing waste). This marks the beginning of AI managing its own workflow, more rigorously than human managers. The article emphasizes that an honest evaluation system (eval) is key to avoiding self-deception.
@Jackywine: Must not miss this Anthropic blog today on running an AI-native engineering organization https://claude.com/blog/running-an-ai-native-engineering-org… Also, does anyone know of a better translation model?
Anthropic's engineering blog discusses how running an AI-native engineering org requires rethinking planning, context gathering, and code review, with agentic coding shifting bottlenecks from coding to verification and review.
@sodawhite_dev: https://x.com/sodawhite_dev/status/2067413032544940062
The article analyzes Anthropic's 400,000-session report on Claude Code, pointing out that AI programming tools are changing the division of labor between humans and AI. Domain knowledge is more important than coding ability. Expert users can enable AI to perform more complex tasks, while verification and task decomposition capabilities become core competitive advantages.
@CMGS1988: https://x.com/CMGS1988/status/2074488576356876585
The author shares personal insights on the relationship between AI and infrastructure, arguing that while AI amplifies execution capabilities, it cannot replace human judgment on complex trade-offs and organizational issues. The author illustrates this with their own experience developing the cocoon/sandbox project using models like Fable 5.
@ba_niu80557: https://x.com/ba_niu80557/status/2068751230667755859
The article explores how increasingly powerful AI models eliminate those whose skills can be encoded into prompts, emphasizing that the truly irreplaceable value lies in tacit knowledge, physical-world operations, and interpersonal trust. Through the example of a friend transitioning from a consultant to a hardware integrator, the author illustrates how proactively yielding to AI-replaceable tasks while deepening expertise in areas beyond AI's reach is key to surviving and thriving in the technological wave.