Skip to content
View RomanSlack's full-sized avatar
👁️‍🗨️
Activating some neurons
👁️‍🗨️
Activating some neurons

Organizations

@RIT-AI-Club @GioiaInc

Block or report RomanSlack

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. Markdown is supported. 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
RomanSlack/README.md

Roman Slack

Lead AI Platform Engineer @ Just-Tech LLC | RL addict | Matrix Emulator
Simulation maximalist. Startup survivor. Halfstacker turned Fullstacker


"Experience"

  • Currently leading AI product development in the legal services sector @ Just-Tech LLC. https://jtai.law
  • Researcher in the missile intercept trenches, integrating LLMs into Unity-based simulated environments.
  • Former Lead AI Engineer at Levangie Labs, led end-to-end development and scaling of AI systems in a fast-paced startup.
  • Conducted applied AI research at the Air Force Research Laboratory Information Directorate.
  • Project manager at the RIT AI Club, overseeing multi-agent research and development initiatives in AI education.

Currently Learning

  • Near-Telepathy using (SSR) silent speech recognition, (Look it up very cool stuff)
  • How to form a better relationship with Claude Code.
  • Architectural Design
  • One of those little bipedal wheel robots but it can transition from wheels to feet.
  • Aura coding (vibe coding is obsolete)
  • C+++ and Python 2

Currently Building For Fun - The Gym

The experiment: can a person and an AI picture the same building and get it to buildable quality purely through talking? So far that's a luxury Nordic gym built into a forest ravine in Elverum, Norway, designed entirely through conversation with Claude. Four floors step down a rock gorge to below the river: a lounge bridge with a walk-on glass floor over the water, a waist-high river window, round oak locker rooms that glow cloudberry, moss and fjord, a stone grotto shower, and little oak cleaning robots called Vaskebjørn. Every piece of furniture is generated from code (no asset packs), the building is constructed stage by stage like a real crew would, and every version can be replayed with the prompt that caused it.

romanslack.com
linkedin.com/in/roman-slack-a91a6a266
https://www.youtube.com/@roman-slack

Pinned Loading

  1. jev-drone jev-drone Public

    Camera-only autonomous drone in MuJoCo with a small judgment model (TypeSafe Jev) in the loop at 2.5Hz

    Python 277 39

  2. Discerio Discerio Public

    Learn Agentic AI through play. Built at HackPrinceton Fall 2025, where it won 🥇1st Place in the YC Challenge and 🥇1st Place in the Dedalus Special Track, Discer.io is an MMO sandbox to build and de…

    JavaScript 2

  3. Hlynr_Intercept Hlynr_Intercept Public

    Hlynr is a simulated reinforcement learning environment for testing fast-response intercept logic. It enables real-time decision-making and trajectory planning under time pressure, designed purely …

    Python 5 2

  4. thought-to-text thought-to-text Public

    A viral edit decompiled frame by frame and rebuilt with code: Remotion visuals, SuperCollider score, ElevenLabs VO

    TypeScript 7

  5. SimuVerse_Framework_MK3 SimuVerse_Framework_MK3 Public

    Model-agnostic Python backend for SimuVerse, enabling agent decision-making, memory recall, and environment interaction in multi-agent simulations integrated with Unity.

    Python 4

  6. AUS-Lab AUS-Lab Public

    AUS-Lab is an experimental framework for LLM-driven cognitive control of UAV swarms. It links natural language reasoning with swarm simulation to enable autonomous coordination, adaptive behavior, …

    Python 4 1