Ihor VItenko

Multimodal AI, Post-Training Recipes, Long-Horizon Agentic Tasks, RL Environments, Agent Interpretability, Synthetic Data Generation, Hybrid Transformer Architectures