Molochynskyi Yehor, R&D Lab
Math & AI
Turn thousands of disconnected company documents into an explorable knowledge graph — making implicit organizational knowledge explicit and navigable.
A library for automatic supervised training of neural network models — handles training loops, logging, checkpointing, and hyperparameter scheduling so you can focus on architecture.
What failure modes appear in GraphRAG retrieval when the underlying graph has noise?
Can graph topology fix embedding collisions during entity resolution?