Engineering
AI Applications & Backend Systems
I build production-oriented AI agents, retrieval systems, and high-concurrency backend services. I care about clear system boundaries, observable workflows, and measurable improvements.
Internships
2026.02–2026.06, AsiaInfo Security, Nanjing, China.
2026.06–2026.09, Institute of Software, Chinese Academy of Sciences, Beijing, China.
Technical Toolkit
JavaSpring BootSpring CloudSpring AILangChain4jRAGAgentsMySQLRedisKafkaRocketMQClickHouseElasticsearchMilvusDockerKubernetes
Selected Projects
Core Developer · 2025.12–2026.02
AIOps Agent Platform
Combined retrieval-augmented generation, agent tools, and automated diagnostics into an end-to-end workflow that reduced troubleshooting time from hours to minutes.
- Designed a Markdown-driven tool registration and loading mechanism.
- Implemented persistent agent state and memory with Redis.
- Streamed model output, tool progress, and PDF reports through SSE.
- Improved retrieval with semantic chunking, embeddings, and reranking.
Backend Developer · 2025.07–2025.12
Flash-Sale System Optimization
Improved consistency and throughput for a high-demand ordering system, supporting peak traffic above 2,000 TPS.
- Used Redis and Lua for atomic inventory deduction and overselling prevention.
- Added Redisson distributed locks to guarantee idempotent ordering.
- Built a local-cache and Redis two-level caching strategy.
- Applied token-bucket rate limiting and RocketMQ for traffic smoothing.