Molochynskyi Yehor, R&D Lab

Molochynskyi Yehor

Math & AI

  • Home
  • Projects
  • Demos
  • Investigations
  • Graph
  • About

Featured Projects

Organizational Knowledge Graph

Turn thousands of disconnected company documents into an explorable knowledge graph — making implicit organizational knowledge explicit and navigable.

active
knowledge-graphs NLP enterprise-AI Python Neo4j

NN Trainer Library

A library for automatic supervised training of neural network models — handles training loops, logging, checkpointing, and hyperparameter scheduling so you can focus on architecture.

completed
deep-learning training-infrastructure automation Python PyTorch

Recent Investigations

active Nov 2024

GraphRAG Retrieval Failure Modes

What failure modes appear in GraphRAG retrieval when the underlying graph has noise?

knowledge-graphs GraphRAG retrieval
completed Oct 2024

Graph Context for Embedding Collision Resolution

Can graph topology fix embedding collisions during entity resolution?

knowledge-graphs entity-resolution embeddings

Demos

Particle Storm live

Particle Storm

65,536 GPU-simulated particles driven by 3D curl noise, collapsing into mathematical forms — Lorenz attractor, Möbius strip, neural network diagrams, runes.

visualization simulation mathematics
Explore the knowledge graph →