AI / NLP Engineer — GenAI & RAG systems
I design, build and evaluate LLM applications: retrieval-augmented generation, agents and graph-backed retrieval. I come from a data science and NLP background (7 years in NLP and machine learning, M.Sc. in Computer Science, thesis on Siamese BERT) and choose architectures by measured results.
| Project | What it is | Status |
|---|---|---|
| Local PDF RAG System | Local RAG pipeline for grounded questions about PDF books: section-aware chunking, FAISS retrieval, answers with citations from a local Ollama model. Retrieval evaluated on a 68-question golden set. | Completed |
| ReAct Agent From Scratch | ReAct agent in plain Python without frameworks. One chat function for Anthropic, OpenAI and local Ollama models; Thought → Action → Observation loop. | In progress |
Next: a LangChain rebuild of the agent for comparison, and a GraphRAG project.
LLM / GenAI: RAG, AI agents, MCP servers, LangChain, prompt engineering and optimization, Ollama Retrieval & data: FAISS, SentenceTransformers, Neo4j, SQL NLP / ML: PyTorch, scikit-learn, Transformers (BERT, SBERT), time-series forecasting Engineering: Python, FastAPI, Docker, GCP, Git, Linux
AI / LLM engineering roles and collaborations on RAG, agents and retrieval systems. The quickest way to reach me is email or LinkedIn.