notadamking

Projects

A selection of things I've built — from open-source reinforcement learning frameworks to AI-powered developer tools.

Active

AI-powered documentation that auto-generates and syncs docs from your codebase.

Analyzes your codebase and updates docs automatically via GitHub Actions on every commit and PR. Block-based editor, OpenAPI references, branch previews, and one-click edge deployment.

Documentation AI GitHub Actions SaaS Open Source
Active

stoneforge

View on GitHub 80+ stars

Multi-agent orchestration platform for running teams of AI coding agents in parallel.

Run Claude Code, Codex, and OpenCode simultaneously with Git worktree isolation, dependency-aware task dispatch, automatic merge review, and a real-time web dashboard.

TypeScript AI Agents CLI Open Source

Rovela

No longer maintaining or involved in this project

AI-powered platform that transforms business ideas into fully functional e-commerce stores.

Enables entrepreneurs to build professional online stores without technical expertise. AI generates store components, design, and functionality from concept through deployment.

AI E-Commerce Next.js SaaS

GameCraft

No longer maintaining or involved in this project

AI-powered toolkit that helps developers ship Unity games 5x faster.

Generates scripts, assets, and playable scenes using AI with full project awareness. Color-coded diffs for team collaboration, with support for Claude, Gemini, DeepSeek, and OpenAI.

Unity Game Dev AI SaaS

TensorTrade

6k+ stars
No longer maintaining or involved in this project

An open-source reinforcement learning framework for training, evaluating, and deploying robust trading agents.

Composable components for building custom RL-based algorithmic trading systems — environments, action schemes, reward functions, and data feeds. Supports training with Ray/RLlib.

Python Reinforcement Learning Trading Open Source

RLTrader

1.8k+ stars
No longer maintaining or involved in this project

A deep reinforcement learning agent for cryptocurrency trading using OpenAI Gym.

The predecessor to TensorTrade. Uses feature engineering, statistical modeling, and Bayesian optimization to train trading agents from historical data. Featured in viral Towards Data Science articles.

Python Deep RL Crypto Open Source