Skip to content
View misza222's full-sized avatar

Block or report misza222

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
misza222/README.md

Hi, I'm Michał 👋

AI Engineer in Warsaw, building LLM-powered applications and agentic systems.

8+ years across ML, VFX production, enterprise data, and software architecture. Now focused on context engineering, agentic AI, and bringing research into production.

Current projects

🧠 llm_coach — AI coaching app with multi-layer memory (working/episodic/semantic), structured CoT via Instructor, LLM-as-Judge evaluation (14 criteria), OAuth, pluggable persistence. Python, FastAPI, React.

🔍 choose-model — CLI that fetches models from OpenRouter, filters by price, benchmarks for latency & capability across 3 tiers. Rust.

🔄 simpla_loop — Educational framework for agentic loop patterns (ReAct, Plan-and-Solve, Reflexion). Clean abstractions, full type annotations. Python.

Background

  • Senior ML Engineer @ Pixomondo (2023–2025) — NeRFs & Gaussian Splatting for VFX
  • Data Engineer @ Boehringer Ingelheim (2018–2022) — ETL, forecasting, automated reporting
  • Software Architect (2012–2017) — APIs, ML solutions; founded Ordys (SaaS, 8 years)
  • PGCert Software Engineering — University of Oxford

Links

LinkedIn · Blog · misza222@gmail.com

Pinned Loading

  1. llm_coach llm_coach Public

    Python

  2. simpla_loop simpla_loop Public

    Simple Agentic loop

    Python

  3. choose-model choose-model Public

    Rust

  4. agent_vis agent_vis Public

    Python