Hi, I'm Artem Legotin

I’ve been a software engineer since 2013, building AI products since 2020.

I’m based in Amsterdam.

What I do

I focus on agentic LLM systems, generative AI, and recommender systems.

Currently, I’m Head of AI at exactly.ai.

Outcome-aware context allocation: testing whether a learned controller can choose how much evidence a RAG system reads, with answer and citation checks.
AffineFlow: learning how a robot responds over time, with a small control-cost gain in simulation.
FamilyTiles: 20% less weight storage for two related language models, with exact reconstruction and measured inference limits.
ExactKV: lossless LLM cache compression, showing when recovery costs more memory than compression saves.
Dependency-aware verification: an experiment in verifying AI-written code without counting the same evidence twice.
Relational Belief Compilation: keeping relationships between uncertain facts so decisions can show when the evidence isn’t enough.
Customer Pilot Evidence Pack: turns search-pilot results into auditable reports, tracking failed requests and checking whether the evidence supports a comparison.

"EML After the Theorem"
A follow-up implementation of the paper "All elementary functions from a single binary operator" that searches data for simple, readable formulas by training and verifying expression trees built from one basic operation.