"Winners don't make excuses when the other side plays the game."
— Harvey Specter
I build intelligent systems that combine data, design, and reasoning.
My work spans from AI-driven tutoring and retrieval systems to scalable data pipelines, always focused on clarity, usability, and measurable impact.
Core Focus
Retrieval-Augmented Generation (RAG) • Agentic Workflows • NLP Systems • Knowledge Graphs • Data Pipelines
Technical Stack
Python LangChain Neo4j FastAPI Hugging Face PyTorch Pandas Scikit-learn Google Cloud BigQuery Looker Studio
Approach
I design reliable, context-aware, and production-ready AI applications, combining experimentation with solid engineering practices.
- Agentic Knowledge Graph Construction for RAG : built a multi-agent system for automated graph creation and refinement, improving factual consistency and retrieval efficiency.
- AI-Powered Tutoring Platform : designed a system that restructures educational materials into session-based learning paths with summaries, quizzes, flashcards, and mind maps.
- Retrieval-Augmented Generation System : implemented a custom RAG pipeline with context tracking and validation to enhance grounded responses.
- End-to-End Big Data Pipeline on Google Cloud : deployed a real-time event-driven ETL architecture with Cloud Functions, Dataflow, BigQuery, and Looker Studio dashboards.
Engineering clarity from complexity.