Mujoco jobs

Job Title: MuJoCo Task Developer for LLM Reinforcement Learning Dataset We are looking for an experienced MuJoCo/Robotics Engineer to create high-quality MuJoCo benchmark tasks for LLM reinforcement learning. Responsibilities: Design MuJoCo tasks (MJCF/XML, morphology, controllers, or debugging tasks) Write clear task instructions () Implement deterministic automatic graders () Create reference solutions and baseline solutions Ensure tasks are reproducible, verifiable, and follow the provided guidelines Requirements: Strong experience with MuJoCo and MJCF Python programming skills Knowledge of robotics simulation and kinematics/dynamics Experience designing evaluation scripts or automated testing Ability to create deterministic, code-verifiable scori...

Job Title: MuJoCo Task Developer for LLM Reinforcement Learning Dataset We are looking for an experienced MuJoCo/Robotics Engineer to create high-quality MuJoCo benchmark tasks for LLM reinforcement learning. Responsibilities: Design MuJoCo tasks (MJCF/XML, morphology, controllers, or debugging tasks) Write clear task instructions () Implement deterministic automatic graders () Create reference solutions and baseline solutions Ensure tasks are reproducible, verifiable, and follow the provided guidelines Requirements: Strong experience with MuJoCo and MJCF Python programming skills Knowledge of robotics simulation and kinematics/dynamics Experience designing evaluation scripts or automated testing Ability to create deterministic, code-verifiable scori...

A MuJoCo task in this repo is a robotics/simulation benchmark where the agent must produce a verifiable artifact, usually an MJCF model file like /tmp/output/ or a controller like /tmp/output/policy.py. The artifact is graded by scorer/ using deterministic MuJoCo checks: compile the model, inspect joints/bodies/sensors, run fixed rollouts, measure stability/control/objective success, and return a score from 0 to 1. The task package lives under problems/<task_id>/ and includes the public prompt (), task metadata/config (), public data/helpers (data/), hidden grader logic/data (scorer/), oracle solution (solution/), and for MuJoCo, a required reviewer video generated by solution/render.sh. The oracle must score exactly 1.0, and the video

...simulation pipelines, and collaborate on robotics systems that enable intelligent robots to learn and operate effectively in complex environments. This is a fully remote contract opportunity with flexible working hours and premium compensation. Key Responsibilities • Design and implement accurate robot models using URDF, MJCF, and SDF formats • Develop and maintain simulation environments using: MuJoCo NVIDIA Isaac Sim Gazebo Isaac Gym (preferred) • Build scalable simulation pipelines for: Robot training Validation testing Motion planning Navigation Manipulation tasks Reinforcement learning workflows • Tune and optimize simulation parameters including: Friction Damping Inertia Contact dynamics Actuator behavior Collision models • Integrate robot...

...advanced AI simulation and reinforcement learning projects focused on MuJoCo environments. This is a high-level remote contract opportunity where you will help design, build, and optimize physics-based simulation environments used to train intelligent agents. The work directly contributes to cutting-edge research in robotics, embodied AI, and sim-to-real learning systems. You will be working on tasks involving locomotion, manipulation, multi-agent systems, and reinforcement learning pipelines used to train next-generation AI systems. Compensation $100 – $150 per hour Remote contract role Flexible working hours 10–40 hours per week Long-term extension opportunities based on performance Responsibilities Design and develop MuJoCo simulation environments for rob...

...limits and look convincingly effortless when played in real time In addition, rename and organize each joint in the Blender file to match the JOINT names on the right side, following the same hierarchical structure. The base file is ready to send as soon as we agree on the workflow. 7 days to finish p.S. Refer to the joint angles for each pose written in radians (RAD) on the right side of the MuJoCo simulator, and reproduce them as closely as possible in Blender. ...

...performed. The following gait animations will be implemented: WALK PACE TROT Canter Galloping ref: Contestants must use the provided or uploaded .obj files of the 3D model. download Duvee zip file in the FILES. All joint motion ranges (range) and physical limits (limit) are specified in the corresponding XML files, which can be opened and reviewed in the MuJoCo Simulator. The Duvee model used in this competition incorporates a Calf structure that is structurally different from the original Unitree Go2 Pro model; accordingly, the bending direction of the knee joint also differs from that of the original Go2 Pro. Contestants must fully understand and account for these structural differences when designing model control logic and animations. Any submission

Please Note: Code it already written, it just needs few tweaks in 12 hour deadline or so. I need help improving a Python MuJoCo pick-and-place pipeline using a UR5e + Robotiq 2F-85 gripper (from mujoco_menagerie). The task: pick a red box and place it in a green bin while avoiding obstacles. The stack includes MuJoCo, ur-analytic-ik (analytical IK), and OMPL (currently RRTConnect) for collision-free motion planning, plus a joint-space PID torque controller and a state-machine execution policy. Your job: tune PID gains (including “loaded” gains when holding the box), improve OMPL parameters (setRange, checking resolution, goal threshold/tolerance, interpolation) and retry/replan logic, and fix any collision/exclusion edge cases (gripper↔box, box↔table/b...

...Computer Graphics, Digital Art, Game Development, or related technical/artistic discipline 1+ years of experience as a Technical Artist in game development, simulation, VFX, or real-time 3D applications Proficiency in 3D modeling software: Maya, Blender, or 3ds Max with strong skills in modeling, UV unwrapping, and texturing Isaac Sim experience or equivalent robotics simulation platforms (Gazebo, MuJoCo, Unity, Unreal Engine) Python scripting proficiency for tool development, automation, and pipeline creation What We Offer Direct impact on the future of robotics simulation, working with cutting-edge AI-driven scenario generation technology Technical ownership and rapid growth opportunities within a venture-backed startup founded by proven IIT Bombay and Microsoft Research ...

About the job About Tonks...environments for visual fidelity, physical realism, computational performance, and cloud deployment. Support customers in integrating their robots and workflows on our platform through direct engineering collaboration. Requirements Bachelor’s degree in Computer Science, Mechanical Engineering, Robotics, or a closely related discipline. Minimum 1 year of hands-on experience in Isaac Sim, Gazebo, MuJoCo, or robotics simulation software. Proficiency in Python and List of related projects and portfolio we could see. What We Offer Chance to build the future of robotics simulation, working directly alongside proven founders from IIT and Microsoft Research backgrounds. Technical ownership and rapid growth opportunities within a Venture-backed, impact-d...

...project involving a quadruped robot, and I need a seasoned professional with a strong expertise in ROS2 control. Key Responsibilities: - Develop and implement ROS2 control systems for the quadruped robot. - Ensure the control systems are optimal for outdoor terrains. Mujoco Simulation: - Design the Mujoco simulation to incorporate realistic physics dynamics. - Ensure the simulation accurately reflects the robot's interaction with outdoor environments. Ideal Skills: - Extensive experience with ROS2 control. - Proficiency in Mujoco simulation, particularly with realistic physics dynamics. - Background in robotics, specifically with quadruped robots. Your expertise will be pivotal in the success of this project. Looking forward to your proposals. Find the pro...

I need an expert in robotics and simulation to set up inverse kinematics for a two-arm system in a MuJoCo environment. The job involves working with a provided XML file describing the arms and ensuring the inverse kinematics function runs in real-time, specifically in less than 10 milliseconds. Key Requirements: - Proficiency in inverse kinematics and its application in robotic simulations - Experience with MuJoCo environment - Ability to optimize the function for real-time performance - Familiarity with mujoco XML files in the context of simulation

...minor debugging and package install, the code should run. Here's what I'd like help with: 1. Simplification. I made this for my research. Thus there are likely portions that could be cut if it's a general environment. Also open to reorganization of the environment if there's a better way to accomplish. 2. "Branding." My plan is to make RL with high performance aircraft as accessible as Atari, Mujoco, etc. Typically vendors hold simulation efforts closed source. Not looking to make a marketing campaign necessarily, but would like your thoughts on how to best reach the audience, aviation enthusiasts. 3. CI/CD. Hoping to publish this on gym as well as push to github in a docker container. Could use guidance on that process and a plan for CI/CD movi...

Control of a Ball on a Beam in MuJoCo. Develop mujoco code for the above problem.

...run in real-time or nearly real-time. We're particularly interested in simulating tool deflections from an environment that can be rigid or compliant. There are a number of physics engines that might be able to solve this problem for us. I’ve leaned more towards the robot simulations that are designed for reinforcement learning because they tend to be faster (Mujoco, Nvidia Gym, Pybullet, Gazebo, etc.). Of this list, it seems that Mujoco does the best with lots of contacts, it’s faster than most of the others, and it’s open source (see papers below). The more realistic direction would be the finite element solutions ([SOFA]() — although it seems that SOFA includes rigid contact and at least mentions a real time emphasis). Write your key skill...

Hello Masoud, I was wondering if you could do some modifications to my MuJoCo environment and model? Regards Marten

Hi Mayur R., I noticed your profile and would like to offer you my project. We can discuss any details over chat.I m working on mujoco to apply optimal control on a kuka iiwa 14 robot while throwing a ball for a specific distance can you help me.

...simulation software called MuJoCo and another one for ROS/Gazebo of the same model. Atlas as a robot model already exist in a previous version for both MuJoCo and ROS, I want to update it to the latest version. Both these seperated models should have one base model with the half ball hands and one with sandia hands (human like hands) attached to it. I want the models to be created as real as possible compared to the real Atlas robot. With amount of movement, and adding sensors and cameras like the real robot have. And as much as possible adding textures etc. Atlas new model (i will provide this one): I found a bunch of great resources to get started below: Building a robot for MuJoCo

UR5 arm simulation in Mujoco, including IK controller accessible with python code

Looking for someone who can help in the robotic simulation using Mujoco or any other dynamics robotics simulation tools

UR5 arm simulation in Mujoco, including IK controller accessible with python code

I need you to develop some code for me. using OPENAI GYM and mujoco using Python. In openai gym for UR5

I need you to develop some code for me .using OPENAI GYM and mujoco using Python. In openai gym for HandManipulateBlock-v0

hello i need someone who can use any algorithm to train humanoid using mujuco