Full-Stack Python Development | FastAPI, PostgreSQL, LLM Fine-Tuning & RAG Systems
I build backend systems and AI-powered applications using Python, with a focus on clean API design and hands-on experience working with LLMs.
What I offer:
Backend & Web Apps
- REST APIs with FastAPI, using Pydantic for request/response validation
- Database design with PostgreSQL + SQLAlchemy, including migrations with Alembic
- JWT authentication and basic user management (login, permissions, CRUD endpoints)
- Async API development for better performance
LLM Fine-Tuning
- Fine-tuning small/quantized language models (e.g., Qwen, GPT-2) using LoRA/PEFT
- Preparing training datasets, including converting raw data into instruction-style training pairs
- Working with 4-bit/8-bit quantization to train on limited hardware (Kaggle/Colab)
RAG (Retrieval-Augmented Generation) Systems
- Building basic RAG pipelines: embeddings, vector storage (pgvector), and retrieval-based generation
- Implementing similarity search for context retrieval
- Connecting local or hosted LLMs to retrieval systems