👋 About Me
I am a Data Scientist with a double Master’s degree (Data Science & Engineering Sciences in DS), currently deepening my understanding of modern AI systems. I’m slowly learning the nuances of RAG, LLMs, Neo4j, vector databases, MCP, and knowledge-augmented retrieval to build grounded, production-ready projects.
🚀 I’m currently working on
LLM-based projects to get comfortable with agentic workflows, tool use, and smart retrieval
Building small end-to-end systems that combine Python + RAG + embeddings
Exploring how knowledge graphs & reasoning can improve AI decision making
🤝 I’m looking to collaborate on
Open-source projects involving RAG, LLM apps, multi-agent systems, or graph + AI
Building practical tools for learning, research, or workflow automation
Anything that makes complex AI concepts easier for beginners
📚 I’m currently learning
Vector search & embeddings
Neo4j + Cypher for knowledge-graph-driven AI
Multi-context retrieval for LLM applications
MCP (Model Context Protocol) and how tools talk to AI
🧠 Would love advice on
Improving parameter efficiency — when to fine-tune, when to adapt, and when not to touch the weights at all (LoRA, QLoRA, adapters, etc.)
Understanding when not to use an LLM — choosing the right tool for the problem instead of forcing generative solutions
Best practices for prompt orchestration across tools — designing structured prompts for multi-step reasoning, tool calling, and agent workflows
Improving RAG retrieval accuracy — context pruning, chunking strategies, vector store choices, and “signal over noise” retrieval
😄 Fun fact
I still open StackOverflow like it’s Google — and somehow, it works every time. 🛠️
