Your local AI engineering assistant
Making engineering, building, making, and tinkering more accessible & fun.
Runs entirely on your machine. Your data stays yours.
Works with your favorite tools
Whether you're designing parts in Blender, slicing prints for your Bambu Lab, writing Arduino code, or just figuring out the right bolt size — Moose is there to help.
An AI assistant that understands engineering, respects your privacy, and runs on your hardware.
Works with Bambu Lab printers, Blender, FreeCAD, and other maker tools.
Your projects and data never leave your machine. Complete privacy.
Use LM Studio, Ollama, or llama.cpp. Pick the AI models you trust.
Script repetitive tasks, batch process files, and automate your workflow.
Material specs, tolerances, unit conversions, and technical calculations.
Get explanations, tutorials, and guidance while you work on projects.
Different problems need different expertise. Moose routes your questions to the right specialist.
Query Routing
Fast-path routing that categorizes queries to get you answers quickly.
Deep Reasoning
Handles complex analysis and multi-step problem solving for tough engineering challenges.
Code & Scripts
Writes code, automates workflows, and helps debug your projects.
Calculations
Engineering math, unit conversions, material calculations, and data analysis.
Planning
Breaks down complex builds into actionable steps and helps you plan projects.
Safety First
Keeps your data local and ensures safe operations throughout.
Get Moose running on your machine. Requires Python 3.11+, Node 18+, and an LLM backend.
git clone https://github.com/moosefinch/moose.git
cd moosecd scripts
python setup.pyAuto-detects your LLM backend and configures everything.
cd backend
pip install -r requirements.txt
./start.shhttp://localhost:3000cp profile.yaml.example profile.yaml
# Edit profile.yaml with your LLM backend settings
docker compose up --buildJoin makers and engineers who want an AI assistant that respects their privacy and runs locally.