Skip to content
View B0yko's full-sized avatar

Block or report B0yko

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
B0yko/README.md

Hi, I'm Andrii Boiko

I lead an AI engineering team in Berlin. We build agents that people rely on at work. In my research, I look at what you can verify when all an agent gives you is its account of what happened.

Selected work

AI Reliability Layer
I built a shared system that spots failures across our production agents and recovers failed runs.

Agent Cheat Index
Did the agent fix the code or change the test? Compare 10 coding-agent setups across 1,162 runs.

Receipts, Not Transcripts
In a team study, we compared agents’ logs with the server record. Their logs left out 573 of 8,211 operations.


Code review: vLLM Semantic Router. I flagged duplicate score IDs; the author added validation and regression tests.

Pinned Loading

  1. tern tern Public

    Search a video archive by what was said, what was on screen, or what the scene looked like.

    Python 193 1

  2. taskdistill taskdistill Public

    Train a small local model for one recurring task and measure when to fall back to the original API.

    Python 3

  3. local-enough local-enough Public

    Measure whether a back-office AI task can run on your own hardware: accuracy vs gold, p50/p95 latency, $/1k tasks, amortised hardware cost and break-even volume, plus an OpenAI-compatible router th…

    Python

  4. pgwarden pgwarden Public

    Give AI assistants access to Postgres with per-user permissions and human approval for writes.

    Python 23

  5. agent-claimcheck agent-claimcheck Public

    Checks an agent’s success claims against its tool calls and the state it left behind.

    Python 31

  6. booking-truth booking-truth Public

    Checks whether a booking agent actually changed the calendar it claims to have updated.

    Python