I'm Steven-Zhl, a passionate fullstack developer from China. I enjoy creating web applications and exploring new technologies in the field of AI. I'm always eager to learn and grow as a developer, and I'm excited to share my projects and experiences with you!
name: Steven-Zhl
pronouns: he/him
born: 2002
education:
major: Computer Science (AI)
degree: Bachelor
language: [Chinese, English]
job: Fullstack Developer (Vue.js & FastAPI)
⚠️ Work keeps me busy, so my activity might be low. More active during holidays!Rest assured, I’ll respond to all messages as quickly as I can.
I firmly believe that tech stacks should be chosen based on actual needs, not just trends. Below is a breakdown of my expertise, categorized by the core scenarios I specialize in.
- Service:
FastAPI, I like asynchronous programming for better performance and scalability.
Pydantic, it helps me build reliable services.
MongoDB, it is better suited for JSON-based memory.
Redis, useful for caching hot data, request rate limiting, simple message queue (based on
Arq), and many other use cases.
- Agent-related:
LangChain (v1.0), Despite many criticizing its over-engineering, I still think it's very insightful.
- Qdrant, a practical vector DB with a concise Python API. I use it to build RAG pipelines, semantic-based search, and recommendation.
- Dify, to build isolated and simple requirements and integrate them into the Agent.
- Frontend:
- Task Queue:
RabbitMQ (Distributed Message Broker)
- Basic Scraping:
httpx,BeautifulSoup4,re - Dynamic Rendering:
playwright(Prioritize protocol-level scraping for performance; use Playwright as fallback.) - Related libraries for stealth features, proxy pool management, captcha solving, etc.
- Data Pipeline:
MongoDB,
Pandas



