| Machine Learning | 70 |
|---|---|
| Computer Vision | 12 |
| Data Science | 65 |
| NLP & LLMs | 66 |
| AI Agents | 58 |
| MLOps | 26 |
Each node is one domain, scored only from evidence in these public repositories. Select one to see why.
Deep, repeated, well-engineered work in this domain.
30 of 51 public repositories contain AI work and are included in this Atlas. The remaining 21 (repositories without AI code, forks and empty repositories) are excluded from scoring and have no effect on the result. How repositories are selected
Ordered by how much each project adds to the Atlas score, not by its own score. A project with evidence in several domains can rank above one with a slightly higher project score. Choose a domain to focus on the projects with meaningful evidence there.
16% of Atlas Score
nexttext is a Python library for debiasing and rebiasing text to reduce gender bias in AI pipelines, providing simple functions to neutralize and reintroduce skew in labels.
14% of Atlas Score
The repository provides a deterministic, no_std 8×8 Minesweeper state machine implemented in Rust as a lightweight reinforcement‑learning environment, complete with observation tensors, action masks, configurable rules, and an extensive test suite.
10% of Atlas Score
The C3PA Explorer is a proof-of-concept web application for exploring the C3PA dataset of annotated privacy policies. It includes a rule-based data pipeline to extract and align text units from HTML source documents, a normalized SQLite database, and a dual-mode frontend (FastAPI backend or browser-based SQL.js) for interactive exploration.
8% of Atlas Score
EuchreGame is a Python library for simulating the card game Euchre, accompanied by a browser-based sandbox powered by Pyodide and a data factory pipeline for generating synthetic training data to distill large language models into Euchre-playing agents.
8% of Atlas Score
The Robot Servo Sequence Designer is a full-stack web application for designing PWM servo motor control sequences for Arduino-based robots. It features a FastAPI backend with SQLite persistence and a static Bootstrap SPA frontend with a visual timeline editor and browser Canvas motion simulator.
7% of Atlas Score
This repository provides an end-to-end system for training, evaluating, and deploying reinforcement learning policies for Minesweeper in a web-based arcade cabinet. It includes a custom game environment, a constraint solver, neural network training scripts, and a Playwright-based test runner for generating showcases.
