B.Sc. Information Engineering @ TUM | Entrepreneur | AI Safety Researcher
Building trustworthy multi-agent systems and privacy-preserving AI.
- Undergraduate at Technical University of Munich (TUM), Information Engineering
- Research interests: AI Safety, Multi-Agent Collaboration, Federated Learning
- Founder & builder — shipping AI-native products from research to production
- Currently exploring cultural safety benchmarks for LLMs and decentralized agent communication
| Project | Description |
|---|---|
| agent-coworker | Let AI agents discover, call, and collaborate across the open internet — E2E encrypted, zero dependencies |
| AgentFax | Decentralized Agent-to-Agent communication infrastructure |
| clawmatch-server | Centralized matching service for multi-agent orchestration |
| CulturalSafeScope | Cultural safety benchmark for LLMs (paper in progress) |
| Zhijie (智阶) | AI-powered learning workbench for university students |
Languages Python · TypeScript · C/C++
AI/ML PyTorch · LLM APIs · Multi-Agent Systems · Federated Learning
Backend FastAPI · PostgreSQL · Redis · WebSocket · SSE
Frontend React · Vite · Tailwind CSS · Framer Motion
Infra Docker · Git · Linux
AI Safety → Cultural safety benchmarks, LLM alignment, trustworthy AI
Multi-Agent → Decentralized communication, agent discovery, collaboration protocols
Federated Learning → Privacy-preserving distributed training
