chongli jin GCGH159

`> whoami`

I am Chongli Jin, an engineer based in Xi'an who builds practical, AI-native software. My current work lives at the intersection of agent systems, retrieval and knowledge graphs, and developer tooling. I enjoy taking an ambiguous engineering problem, finding the smallest useful system, and then iterating in the open until it earns its place in a real workflow.

Principle: Make intelligent systems observable, composable, and useful—not merely impressive in a demo.

Now exploring

Agent harnesses, multi-agent orchestration, coding agents, and systematic agent evaluation.

Building with

Java and Python across Spring AI, FastAPI, LangGraph, LangChain, Neo4j, Docker, and the Model Context Protocol.

`> featured_work`

Discover contribution opportunities, follow PR review status, and claim issues from a focused Python and SQLite CLI. Built for contributors who want signal instead of tab overload. Explore the repository →	A private-domain RAG exploration combining GraphRAG, LightRAG, and Neo4j to move beyond keyword matching toward relationship-aware retrieval. Explore the repository →
Experiments and implementation patterns for building knowledge graphs with LangChain and Neo4j. Explore the repository →	A playful multi-agent environment using LLM players to explore agent interaction, strategy, and emergent behavior. Explore the repository →

Browse all repositories →

`> open_source`

I contribute where I can add durable value: small quality-of-life fixes, documentation that removes friction, and features that make AI developer tools more capable.

Project	Contribution	State
alibaba/open-code-review	Refactored test string helpers	Merged
QwenLM/qwen-code	Managed media storage with GC and upload cache	Open
QoderAI/better-harness	Qwen Code host adapter	Merged
bytedance/deer-flow	Local start guide	Merged

`> stack --current`

`> github --telemetry`

`> contribution_map`

_{Always open to a thoughtful technical conversation or a useful open-source collaboration.}

Navigation Menu