Johnny

Johnny

AI agent builder and USTC undergraduate
#998055716 followers0 following
>10,000All time
6 KP

About

Third-year undergraduate at USTC's School of Artificial Intelligence. I build AI agents and post-training systems, with interests in vision-language models, diffusion language models, and world models. Currently building TreeWork, a tree-guided development plugin for long-running coding agents. My long-term goal is to build a real-world JARVIS.

Badges

Tastemaker
Tastemaker

Forums

TreeWorkp/treework•

16h ago

Can your coding agent recover the project state, not just read the code?

Most coding agents can reopen a repository. That is not the same as resuming a project.

After a long run, context compression, an interruption, or an agent handoff, the agent can inspect the code and retrieve old notes. But it still has to reconstruct questions such as:

What did the user actually accept?

Which part of the project is active, and why?

•

3h ago

TreeWork - Tree-guided development for long-running coding agents

TreeWork is an open-source Codex plugin that gives coding agents a durable map of complex projects. It aligns requirements before coding, organizes work as a branch tree with dependencies, and keeps Spec, Plan, Progress, Findings, Verification, and lifecycle state on each branch. After context compression, interruptions, or handoffs, agents recover accepted project state instead of reconstructing it from code and scattered memories. Includes a CLI, Git worktrees, Project Map, and Web Paper.
View more