Skip to content
View NeoLorenzo's full-sized avatar

Block or report NeoLorenzo

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 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
NeoLorenzo/README.md

Lorenzo Roque Dal Fabbro

I build software around reinforcement learning, simulations, complex systems, and problems I personally want solved.

I'm a Politics & Philosophy student interested in the intersection of AI, economics, institutions, and human behaviour. Most of my work is experimental: I like building systems, observing what emerges, and documenting the process.

Currently focused on PolyVision and Econ-Engine.

Featured projects

A reinforcement-learning research environment for The Battle of Polytopia.

  • PPO agents trained through Gymnasium and PyTorch
  • 5,517 genuine Polytopia maps
  • 63,913 globally stable action IDs with legality masking
  • Reproducible validation, baselines, human benchmarking, and experiment tracking
  • Java game engine connected to Python through Py4J

The current research task focuses on early-game economic decision-making before progressively expanding toward the full game.


A deterministic browser-based agent economic simulation for experimenting with markets, firms, households, inequality, competition, taxation, and government policy.

The project is developed incrementally, with each MVP adding another economic mechanism while preserving inspectability and reproducibility.

Run the simulation


An evidence-aware lifting tracker built around how I actually train.

It combines workout logging with exercise biomechanics, muscle modelling, equipment tracking, estimated strength progression, body-weight data, and an accompanying scientific literature layer.

Open the app


An experimental artificial-life simulation exploring emergent behaviour in populations of autonomous agents.


A personal productivity system built specifically around my own work, projects, writing, and routines.

What I'm interested in

reinforcement learning · game AI · agent-based modelling · economics · complex systems · artificial life · AI-assisted software development

I am particularly interested in projects where relatively simple rules produce behaviour that is difficult to predict in advance.

Elsewhere

Website: lorenzoroque.com

I also write about AI, technology, politics, philosophy, and the things I'm building.

Pinned Loading

  1. PolyVision PolyVision Public

    Polyvision is a reinforcement learning platform for training and benchmarking Polytopia agents through a headless Java engine connected to Python via Py4J. Its first phase targets economy-first pla…

    Python 1

  2. Lorenzos-Lifting-Ledger Lorenzos-Lifting-Ledger Public

    An installable, evidence-aware lifting tracker with a public scientific overview and an authenticated personal training ledger.

    PLpgSQL 1

  3. FabbroFactory FabbroFactory Public

    Personal operating workspace built with Next.js and Supabase. Fabbro Factory combines strategy, projects, tasks, progress signals, and a private scoring lab in one installable web app.

    JavaScript 1

  4. Econ-Engine Econ-Engine Public

    Econ-Engine is a browser-based agent economic simulation built to make every outcome inspectable.

    TypeScript 1

  5. Life-Engine-3D Life-Engine-3D Public

    Life Engine 3D is a Unity simulation where AI-driven virtual humans perceive, remember, and navigate a dynamic world. Using custom behavior trees, agents manage hunger, sleep, temperature, dangers,…

    C# 1