Neo4j
LLM & RAG
Streamlit
Graph-Powered RAG Recommendations
Contextual recommendations with retrieval-augmented generation on a Neo4j graph.
Selected work
“If you can't solve a problem, there is an easier problem—find it and solve it.” — George Pólya
Contextual recommendations with retrieval-augmented generation on a Neo4j graph.
Dynamic prototype with full front- and back-end functionality.
Serverless extraction and BigQuery analytics with monitoring and retries.
C++/GEANT4 simulation of neutron interactions for LEGEND-1000.
Solving PDEs with physics-informed neural networks in PyTorch.
Classifying astronomical events; discriminating gamma signals from noise.
Predicting particle collisions at the LHC with ML approaches.
Modeling stellar evolution and related phenomena using numerics.
Dynamics of celestial bodies with numerical methods in Python.