robotics

harvard-edge/cs249r_book

An open textbook covering ML systems from edge to agentic AI — This is Harvard CS249r's open-source textbook on engineering AI systems, spanning four volumes from foundations and scaling to agentic and physical AI. It's built as living courseware with validation workflows and companion tools like TinyTorch, aimed at students and engineers learning ML systems design across cloud, edge, and embodied platforms.

earthtojake/text-to-cad

A local CAD toolkit that plugs straight into AI coding agents — This plugin lets AI agents like Claude Code, Codex, Cursor, Gemini and Grok generate 3D models (STEP, GLB, STL, 3MF), run design-for-manufacturing checks, and produce engineering drawings, with links to 3D printing, sheet metal and CNC services. Useful for engineers and makers automating CAD work through agent skills rather than manual modeling.

google-deepmind/mujoco

The physics engine behind modern robotics and ML research — MuJoCo is Google DeepMind's open-source physics simulator for articulated bodies, built for fast, accurate contact dynamics. It offers a C API, Python bindings, a Unity plug-in, and browser-based visualization, making it a core tool for robotics, biomechanics, and reinforcement learning research.

kornia/kornia

Kornia aims to be the spec for differentiable vision in PyTorch — Kornia is a PyTorch-based differentiable computer vision library offering batch, GPU-accelerated, auto-differentiable image processing and geometric algorithms. Its stated direction is to become the reference implementation and executable specification for differentiable CV and geometry in the PyTorch ecosystem, useful for researchers and engineers building spatial AI, augmentation pipelines or robotics vision.

rerun-io/rerun

A columnar data layer built for multimodal robotics logging — Rerun ingests multi-rate robotics data—images, point clouds, transforms, time series, video—from logs, sim, or MCAP/LeRobot formats, then lets you visualize it synchronously in a built-in viewer, query it as dataframes or SQL, and stream it into training. Built in Rust with Python, Rust, and C++ SDKs, it targets robotics and physical AI teams handling heterogeneous sensor data.

isaac-sim/IsaacLab

GPU-accelerated robot learning framework now spans multiple physics backends — Isaac Lab 3.0.0 adds multi-backend support (Newton, OVRTX, OVPhysX, Isaac Sim) for physics and rendering, aiming for accurate sim-to-real transfer. It ships 16+ robot models, 30+ ready-to-train RL environments compatible with RSL RL, SKRL, RL Games, or Stable Baselines, plus multi-agent RL support, useful for robotics researchers needing fast, flexible simulation.

google-deepmind/mujoco_menagerie

Ready-to-simulate robot models for MuJoCo — A curated library of well-tested MuJoCo models — arms, hands, quadrupeds, humanoids and more — designed to work correctly out of the box rather than producing unexpected physics behavior. Now distributed as a PyPI package for easier integration into Python robotics and simulation workflows.

NVIDIA/skills

A verified skill catalog that teaches coding agents NVIDIA's stack — This repo is a continuously-synced catalog of NVIDIA-verified Agent Skills, installable via the skills CLI into Claude Code, Codex, and similar tools. It packages instructions for Physical AI, robotics, simulation, CUDA-X, and RAG workflows, letting agents use NVIDIA software correctly without manual setup or copying skill folders by hand.

KingKongRobotics/jumper

🦀 Jumper — an crab robot.

LuwuDynamics/xgoduck_hardware

XGO-Duck: a 3D-printable biped adapted from Pollen Robotics' Microduck — XGO-Duck is a hardware starting point for a duck-shaped biped robot built around Arduino UNO Q and 15 Feetech servos, adapted from Pollen Robotics' Microduck. It provides mechanical models, PCB designs, a BOM, and an assembly guide, and links to companion repos for onboard runtime and reinforcement-learning experiments. Suited for makers who want to print, assemble, and then explore custom motion.

NVlabs/alpasim

NVIDIA's Modular Closed-Loop Simulator for End-to-End AV Research — AlpaSim is an open-source, Python-based closed-loop simulator from NVlabs for testing end-to-end autonomous driving policies. It pairs pluggable sensor rendering (NuRec, plus video-model rendering via OmniDreams/FlashDreams) with a microservices architecture, letting researchers swap components and scale across nodes for algorithm validation and benchmarking.

joeynyc/awesome-microduck

Tracking the Pre-Launch Ecosystem of Pollen's Open-Source Duck Robot — A curated list for Microduck, the Apache-2.0 biped duck robot from Pollen Robotics and Hugging Face whose walking, kicking and recovery behaviors are MuJoCo-trained ONNX policies. Since hardware ships only around December 2026, this list focuses on what's usable now: simulators, training tools, MCP/agent integrations and community coverage, for roboticists and RL researchers tracking the project ahead of release.

aws-samples/sample-rfm-simulation-on-aws

1-click robot foundation model (VLA, world action model, LLM agent) simulation and evaluation on AWS GPU instances with AWS CDK