Drift
Backed byAntler

Your robotics native agent
grounded in physical world intuition, to accelerate engineering with precision.

Paste this in your terminal to start drifting

Linux / Ubuntu

Used by engineers at 50+ institutes/companies, notably at

Technische Hochschule IngolstadtNortheastern UniversitySingapore University of Technology and DesignGeorgia TechUC BerkeleyOpen DroidsAgile RobotsNeurocadCarnegie Mellon UniversityStanford UniversityUniversity of MarylandIIT RoorkeeJohns Hopkins UniversityUniversity of Wisconsin FoundationUniversity at BuffaloUC San DiegoTechnische Hochschule IngolstadtNortheastern UniversitySingapore University of Technology and DesignGeorgia TechUC BerkeleyOpen DroidsAgile RobotsNeurocadCarnegie Mellon UniversityStanford UniversityUniversity of MarylandIIT RoorkeeJohns Hopkins UniversityUniversity of Wisconsin FoundationUniversity at BuffaloUC San Diego
drift
nikhil@ubuntu-dev: ~/drift-tests

what would you like to do?

drift>

WHY DID WE BUILD DRIFT

We robotics engineers spend weeks wrestling our workspace & simulations.

  • Simulations take 70% of time of an entire robotics engineering pipeline

  • Most of engineering time lost to debugging: setup, broken docs, runtime issues

  • Multiple disconnected tools with no shared context or orchestration

  • Coding agents lack domain expertise in robotics, physics, 3D world understanding

The future of robotics engineering

Your CLI agent for building, controlling anddebugging simulations with just prompts

In this demo, Drift takes a natural-language prompt and scaffolds a production-grade ROS 2 workspace, generates the matching robot and world description files (URDF, SDF, MJCF, or USD), launches the simulation in Gazebo, MuJoCo, or Isaac Sim, runs the requested task, and surfaces debugging output - all from the command line.

DRIFT IS ALL YOU NEED

Robotics workflow orchestration witha domain expert agent

Drift is an intelligent orchestrator across your OS, terminal, ROS, workspace, and simulator simultaneously - through a natural language interface.

Get started
User Prompt
Workspace Flow
Operating System
Processes, env, package & system state.
OBSERVE · EXECUTE
Terminal
Live shell I/O, command monitoring.
CAPTURE · RUN
Workspace
Packages, URDF / SDF / MJCF / USD, launch & config files.
READ · GENERATE / EDIT
Drift
Up-to-date Robotics Expertise3D Spatial ReasoningRoot-Cause Diagnosis
Multi-step Task PlanningPhysics-Aware ReasoningNative Simulator Control
Robotics dev intelligence
Robotics Flow
ROS
Nodes, topics, services, transforms, parameters.
INSPECT · PUBLISH / CALL
Simulators
Gazebo · MuJoCo · Isaac Sim - world building & accurate physics.
OBSERVE · CONTROL
Plugins & Tools
Build systems, visualizers, debuggers, custom integrations.
QUERY · INVOKE
Legend
User intent
Workspace flow
Robotics flow

HOW TO DRIFT

From prompt to your first simulation in minutes

Get Started

001

Start drifting

Run drift in your terminal and describe what you want in natural language - the robot, the world, the specs, the tasks.

002

Drift builds the workspace

Robot description files, world files, launch scripts, the full ROS workspace - generated from scratch or edited in place, in minutes.

003

Sim in action

Drift launches your simulation. Prompt the robot to run tasks, or spin up world variations to stress-test scenarios - all from the same chat.

Features

From prompt to running simulation, in minutes

From a single natural-language prompt, Drift decomposes the goal into ordered subtasks, generates the required code and assets, and executes them end-to-end - planning, replanning, and adapting as the workspace evolves.

drift / output
ALOHA · bimanual peg insertion · mpc converged
drift
~/drift-tests/bimanual

what would you like to do?

drift>
drift / output
unitree go1 · backflip · physics converged
drift
~/drift-tests/quadruped

what would you like to do?

drift>
drift / output
mujoco mpc · 14 joints commanded · controllers live
drift
~/drift-tests/bimanual_ctrl

what would you like to do?

drift>
drift / output
dexterous hand · cube solved · 3/5 converged
drift
~/drift-tests/dexterous

what would you like to do?

drift>

Usecases

Infinite possibilities to use Drift

Testimonials

See what the community is building

Deva Shravan
Deva Shravan

Drift is a really useful AI copilot for robotics. Unlike general AI coding tools, it understands this niche much better and helps automate many boring tasks like ROS package setup and writing world files for simulations. The beta version already works pretty good and I'm excited to see how much better it gets in future versions.

Edwin Huertas
Edwin Huertas

being able to spin up a full robotics sim from a single prompt is prety wild. the fact that it tracks ROS states and workspace in real time makes it way more useable than most sim tools ive tried. congrats on the launch

Alberto Polini
Alberto Polini

I spent several months working on robotic simulations with ROS and Webots, the concept of your tool is very interesting 👍

Piotr Sędzik
Piotr Sędzik

been wrestling with ROS simulation setup for our hardware integrations and this looks like exactly what we needed six months ago. love that it handles the OS orchestration too - that's usually where things get messy. how does it perform with real-time constraints when you're testing control loops?

Aayushi Shrivastava
Aayushi Shrivastava

I used to spend hours setting up simulation environments for my robotics projects. This tool leverages AI to bridge the gap between algorithm design and ROS implementation seamlessly. If you want to prototype faster without the usual setup headaches, this is it.

Pin Z
Pin Z

Really impressive - the idea of prompt-to-simulation for robotics is a huge unlock. The fact that it actively tracks ROS states and fixes issues on the fly is what separates this from just being another wrapper. Curious: do you have plans to support multi-robot swarm simulations? That would be incredible for drone research. Congrats on the launch!

Jason Howie
Jason Howie

Really impressed by how Drift collapses what used to be a multi-day robotics sim setup into a single prompt - building the robot, world, and control loop in one go is the kind of abstraction that unlocks a whole new tier of prototyping speed. Would love to know how it handles sim-to-real transfer gaps when physical dynamics diverge from the model. Feels like a genuine "Cursor for robotics" moment. Congrats on the launch!

Deva Shravan
Deva Shravan

Drift is a really useful AI copilot for robotics. Unlike general AI coding tools, it understands this niche much better and helps automate many boring tasks like ROS package setup and writing world files for simulations. The beta version already works pretty good and I'm excited to see how much better it gets in future versions.

Edwin Huertas
Edwin Huertas

being able to spin up a full robotics sim from a single prompt is prety wild. the fact that it tracks ROS states and workspace in real time makes it way more useable than most sim tools ive tried. congrats on the launch

Alberto Polini
Alberto Polini

I spent several months working on robotic simulations with ROS and Webots, the concept of your tool is very interesting 👍

Piotr Sędzik
Piotr Sędzik

been wrestling with ROS simulation setup for our hardware integrations and this looks like exactly what we needed six months ago. love that it handles the OS orchestration too - that's usually where things get messy. how does it perform with real-time constraints when you're testing control loops?

Aayushi Shrivastava
Aayushi Shrivastava

I used to spend hours setting up simulation environments for my robotics projects. This tool leverages AI to bridge the gap between algorithm design and ROS implementation seamlessly. If you want to prototype faster without the usual setup headaches, this is it.

Pin Z
Pin Z

Really impressive - the idea of prompt-to-simulation for robotics is a huge unlock. The fact that it actively tracks ROS states and fixes issues on the fly is what separates this from just being another wrapper. Curious: do you have plans to support multi-robot swarm simulations? That would be incredible for drone research. Congrats on the launch!

Jason Howie
Jason Howie

Really impressed by how Drift collapses what used to be a multi-day robotics sim setup into a single prompt - building the robot, world, and control loop in one go is the kind of abstraction that unlocks a whole new tier of prototyping speed. Would love to know how it handles sim-to-real transfer gaps when physical dynamics diverge from the model. Feels like a genuine "Cursor for robotics" moment. Congrats on the launch!

FAQs

Frequently Asked Questions

  • Drift is an end-to-end robotics simulation engineering agent. It turns natural-language prompts into production-grade simulation workspaces for ROS2, Gazebo, MuJoCo, and Isaac Sim, generating URDF, SDF, MJCF, and USD files, configuring plugins and controllers, and scaffolding colcon workspaces. Beyond setup, Drift runs scenario variations, captures synthetic data for training, and (coming soon) trains policies on managed cloud compute, collapsing the full sim engineering loop into a single tool.

  • Drift supports ROS2 and three major simulators: Gazebo, MuJoCo, and Isaac Sim (coming soon). Drift generates the matching world and robot description files: SDF for Gazebo, MJCF for MuJoCo, and USD for Isaac Sim. It also produces the launch files, controllers, and plugin configuration needed to run them end to end.

  • General-purpose coding agents don't understand ROS2 plumbing, URDF and SDF semantics, simulator physics, or colcon build conventions. Drift is purpose-built for robotics. It reasons about TF trees, joint controllers, plugin compatibility, mass and inertia validity, and Gazebo, MuJoCo, and Isaac Sim configuration as first-class concepts. The result is working simulations on the first run, not code that compiles but won't launch.

  • Yes. Drift is an end-to-end simulation engineering agent, not just a setup tool. Tell Drift the axes you want to sweep, such as lighting, sensor noise, object placement, friction, obstacle density, payload mass, or starting pose, and it generates parameterized world variants, runs them, and reports where your robot succeeds or fails. This turns simulation from a one-off setup into a repeatable test loop you can iterate on like a CI pipeline.

  • Yes. Drift can scaffold synthetic data pipelines on top of your Gazebo, MuJoCo, or Isaac Sim worlds, capturing RGB, depth, segmentation masks, point clouds, joint states, and force/torque streams in standardized formats like ROS bags and parquet. This makes it straightforward to generate training datasets for perception models, behavior cloning, and policy learning directly from Drift-generated simulations, without writing recording infrastructure yourself.

  • Coming soon. Drift is rolling out training-in-simulation support backed by managed cloud compute, letting you spin up parallel sim instances on GPU servers, run reinforcement learning or behavior-cloning loops, and pull trained policies back to your local workspace, all from the Drift CLI. This closes the full loop inside one tool: world generation → scenario testing → data collection → policy training.

  • Drift combines frontier LLMs with custom-trained agents for robotics tooling and physics-aware models grounded in real-world dynamics. This combination lets Drift catch physically invalid configurations like bad inertia tensors, broken joint limits, or unstable collision geometry before launch, instead of debugging them in simulation afterward.

  • Not yet. Drift today generates simulation workspaces, including URDFs, controllers, launch files, and Gazebo, MuJoCo, and Isaac Sim worlds. The generated ROS2 packages can be ported to real hardware manually, but a first-class sim-to-real deployment flow is on the roadmap, not in the current beta.

  • Yes. Drift accepts existing URDF and SDF files referenced in your prompt (for example, "use my robot.urdf and add a stereo camera") and extends them in place. Drift performs best when existing files follow standard ROS2 conventions; highly customized or non-standard structures may need manual adjustment.

  • Yes. Drift configures standard Gazebo Harmonic plugins automatically, including cameras, LiDAR, IMUs, depth sensors, and force/torque sensors. For third-party or custom .so plugins, mention them by name in your prompt and Drift will add the plugin block to the SDF; you may still need to verify build paths and plugin parameters manually.

  • No. Drift is ROS2-only. ROS1 support is not on the roadmap, since the ROS ecosystem itself is migrating off ROS1 (Noetic reached end of life in May 2025).

  • Drift runs locally on your machine. Your workspace, URDF and SDF files, mesh assets, and ROS code never leave your environment. Only the text of your prompts and the diffs Drift proposes are sent to our API. This makes Drift safe to use on internal robot designs and IP-sensitive workspaces. An air-gapped or offline mode is not available in the current beta.

  • Drift runs on Ubuntu. macOS and Windows are not officially supported today; engineers on those platforms typically run Drift inside an Ubuntu VM, a container, or a remote dev box.

Blogs

Recent highlights.

All posts

Get in touch

We work closely with startups and robotics labs for customized offerings. Drop your email and we'll reach out.