Welcome to my GitHub! I'm a Python-first full-stack developer building production-grade AI applications with a focus on scalable backend systems, async-first architecture, and AI integration. I'm passionate about clean code, type safety, and turning complex problems into elegant solutions.

I specialize in:

• Backend systems: Python async (FastAPI), SQLAlchemy, database design, microservices
• AI/RAG applications: LangChain, vector databases (ChromaDB), semantic search, streaming
• Full-stack development: React + TypeScript frontends paired with robust Python backends
• Cloud-ready architecture: Docker, CI/CD pipelines, production hardening, observability

SourceLogic — A multi-tenant RAG codebase explorer built with:

• Backend: FastAPI (async), SQLAlchemy 2.0, ChromaDB, LangChain, OpenAI integration
• Frontend: React 18, TypeScript, Vite, Tailwind CSS
• Quality: MyPy strict, Ruff formatting, 77% test coverage (138 tests), Bandit security scanning
• Auth: Dev mode, API key (SHA-256), JWT Bearer tokens — fully tested and production-hardened
• DevOps: Docker Compose, Alembic migrations, GitHub Actions CI/CD

This project demonstrates production maturity across authentication, multi-tenancy, testing, security, and deployment readiness.

Languages:

• 🐍 Python (primary) — async/await, type hints, clean architecture
• 🟦 TypeScript (secondary) — React, strict mode, semantic correctness
• Exposure to Java, C#, PHP

Frameworks & Libraries:

• Backend: FastAPI, SQLAlchemy 2.0 async, Pydantic, LangChain
• Frontend: React 18, Vite, Tailwind CSS, Framer Motion
• AI/Data: LangChain, ChromaDB, sentence-transformers, OpenAI API
• Testing: pytest, pytest-asyncio, Vitest, React Testing Library

DevOps & Tools:

• Database: SQLite (dev), PostgreSQL-ready, Alembic migrations
• Infrastructure: Docker, Docker Compose, GitHub Actions
• Code Quality: MyPy strict, Ruff, Bandit, pytest-cov (≥70%)
• Version Control: Git, GitHub

Security:

• Multi-tenant isolation (ORM-level filtering)
• API key authentication (timing-safe comparison)
• JWT Bearer token support
• Rate limiting per tenant
• Input validation (Pydantic)

• Advanced async Python patterns and optimization
• Distributed job queues and worker orchestration
• Observability (metrics, tracing, APM)
• Production Postgres deployment and scaling
• RAG quality optimization for large codebases
• MLOps and model serving

✅ Type-safe by default — MyPy strict mode, Pydantic validation everywhere
✅ Production-minded — Security hardening, testing, error handling, observability
✅ Async-first architecture — Zero blocking operations, scalable by design
✅ AI-integrated — Built RAG pipelines with real streaming and citations
✅ Well-tested code — 77% coverage, E2E auth tests, fixture-based unit tests

I'm always up for discussing software architecture, AI integration, Python best practices, or exploring new tech.

• Email: federico080802@gmail.com
• GitHub: @federico-orsi-dev
• LinkedIn: Federico Orsi (portfolio projects)

Currently building: Production-grade AI applications. Next: distributed job systems and advanced observability.

Open to: Collaborations on Python/AI projects, technical discussions, and opportunities that align with building scalable, well-engineered systems.

Let's build something meaningful together! 🚀