I'm a full-stack engineer passionate about building scalable, maintainable systems with modern architecture patterns. I specialize in Python backend development, TypeScript/JavaScript SDK development, and AI-powered automation systems. My work focuses on domain-driven design, event-driven architectures, and creating developer-friendly tools.
- Languages: Python 3.11+, TypeScript, Go
- Frameworks: FastAPI, Express.js
- Architecture Patterns: DDD (Domain-Driven Design), CQRS, Event Sourcing, Hexagonal Architecture, TAO Pattern
- Async Programming: asyncio, async/await patterns
- Data Validation: Pydantic, Zod
- Logging: Loguru, Winston
- Cloud Platforms: AWS (ECS, S3, CloudFront, DynamoDB), Cloudflare Workers
- Databases: MySQL (aiomysql), Redis, OpenSearch, DynamoDB
- Containerization: Docker, Docker Compose
- CI/CD: AWS ECR, Terraform
- Message Queues: Redis Pub/Sub, Celery
- Languages: TypeScript, JavaScript
- Build Tools: Vite, TypeScript Compiler
- Module Formats: ESM, UMD
- Testing: Vitest, Jest, pytest
- LLM Integration: LangChain, OpenAI, Anthropic, Google Gemini
- Browser Automation: Playwright, Puppeteer
- Sandbox Environments: E2B, Daytona, Docker, Fly.io
I design and implement microservices following DDD principles with clear separation of concerns:
- Domain Layer: Business logic, entities, value objects, domain services
- Application Layer: Use cases, command/query handlers, event handlers
- Infrastructure Layer: Database repositories, external service integrations
- API Layer: FastAPI routers with proper dependency injection
- Message bus implementation for domain events
- CQRS (Command Query Responsibility Segregation) patterns
- Event sourcing for audit trails and state reconstruction
- WebSocket-based real-time event streaming
- Type Safety: Full TypeScript coverage, Pydantic models for runtime validation
- Testing: Unit tests, integration tests, E2E tests with comprehensive coverage
- Documentation: Clear code comments, architecture documentation, API references
- Async-First: All I/O operations use async/await for optimal performance
A microservices platform built with FastAPI following DDD and TAO patterns. Features include:
- Multiple domain services (auth, subs, proto, buzz, nge, link, space)
- Async MySQL connections with connection pooling
- Event-driven architecture with message bus
- CQRS implementation for read/write separation
- OpenSearch integration for full-text search
- Stripe integration for subscription management
- AWS S3 for file storage and CloudFront CDN
Technologies: Python 3.12+, FastAPI, aiomysql, Redis, OpenSearch, Pydantic, Poetry
An AI-powered task automation system with real-time WebSocket communication:
- Long-running async task management
- Multi-sandbox support (E2B, Daytona, Docker, Fly.io)
- Real-time task state streaming via WebSocket
- Multi-LLM support (GPT, Claude, Gemini)
- Browser automation with Playwright
- Distributed architecture on AWS ECS + ELB
- Celery for background task processing
Technologies: Python 3.11+, FastAPI, SocketIO, Celery, Redis, LangChain, Playwright, Poetry
TypeScript/JavaScript SDK for Trickle Backend APIs:
- Full TypeScript type definitions
- Token management with automatic refresh
- Support for both browser and Node.js environments
- SSR support with custom storage
- Comprehensive error handling
- UMD and ESM module formats
- Complete test coverage with Vitest
Technologies: TypeScript, Vite, Vitest
Image processing and OCR service:
- Multi-provider OCR support (Google Cloud Vision, Azure Computer Vision)
- Image segmentation and extraction
- Express.js REST API
- Cloudflare Workers for edge computing
- Sharp for image processing
- Rate limiting and security middleware
Technologies: TypeScript, Express.js, Google Cloud Vision, Azure Computer Vision, Sharp, Cloudflare Workers
An educational project demonstrating the evolution of AI coding assistants from simple to complex:
- 4 progressive versions (v0-v4) showing 16,703% code growth journey (33 → 5,546 lines)
- Modular architecture with event-driven design and sub-agent delegation system
- 15 tools including file operations, search, system commands, conversation management, and delegation
- Intelligent conversation compression with 40-60% token reduction in long conversations
- Sub-agent delegation with 4 specialized agent types and restricted tool sets
- Comprehensive testing with 960+ lines of test coverage
- Complete documentation showing architecture evolution and best practices
- Demonstrates DDD principles, tool system design, and production-ready patterns
Technologies: Python 3.7+, OpenRouter API, DeepSeek v3.2, Modular OOP Architecture
- System Design: Designing scalable microservices with proper separation of concerns
- API Development: RESTful APIs with FastAPI and Express.js, WebSocket real-time communication
- Database Design: Relational database design, async query optimization, search indexing
- SDK Development: Creating developer-friendly SDKs with comprehensive type safety
- AI Integration: Building AI-powered systems with LLM orchestration and tool calling
- DevOps: Docker containerization, AWS cloud infrastructure, CI/CD pipelines
- Code Quality: Writing maintainable, testable code with proper error handling
- Architecture First: Design systems with clear boundaries and responsibilities
- Type Safety: Leverage static typing to catch errors early
- Async by Default: Use async/await for all I/O operations
- Test Coverage: Write tests alongside code development
- Documentation: Keep code and architecture documentation up to date
- Developer Experience: Build tools and SDKs that are intuitive and easy to use
- GitHub: @wupeng
- Projects: Check out my repositories for detailed code examples and architecture decisions
Building robust, scalable systems with modern architecture patterns and best practices.