The Lab

Experiments & tinkering

Smaller things I build for fun and curiosity — AI agents, computer-vision tools, voice models and developer utilities. Each one will get its own write-up over time.

🤖

Sudo AI Assistant

An AI agent that controls an entire Windows PC using tools plus screen vision — it sees the screen and acts on your behalf.

AI AgentComputer VisionPython
🎙️

Voicebox

An open-source voice synthesis studio for generating and shaping synthetic speech.

Voice AITTSOpen Source
🔍

CoC Loot Finder

A Clash of Clans loot detector that reads the screen with OCR and computer vision (EasyOCR / Tesseract) to surface the best targets.

Computer VisionOCRPython
🖌️

AI Canvas Overlay

A transparent on-screen overlay for drawing, annotating and running AI actions over anything on your display.

DesktopOverlayAI
📋

Screenshot to Text

A fast utility that turns any screenshot or screen region into editable text via OCR.

OCRProductivity
🎵

Custom Music Bot

A Discord music bot that plays from YouTube or a local library with simple chat commands.

DiscordBotAudio
📎

Clipboard Manager

An advanced clipboard history manager that keeps both text and images at your fingertips.

DesktopProductivity
🧠

Custom SLM

An experiment in training and running a small language model from scratch.

MLLLMResearch
🗣️

Custom Voice Model

A voice-cloning / custom voice model experiment for personalized speech synthesis.

Voice AIML
🧩

Embedding Server

A standalone microservice that serves text embeddings for search and RAG pipelines.

AI InfraEmbeddingsAPI
📱

Scrcpy GUI

A friendly GUI for screen-mirroring your phone, with unique audio routing that streams phone audio to both the phone and OBS at once.

DesktopStreamingTooling
⬇️

Universal Video Downloader

A multi-platform video downloader for grabbing media from across the web.

PythonUtility

…and many more in progress. Check back as these get documented.