Software Engineer — Controls, Simulation & Dexterous Manipulation

Software Engineer — Controls, Simulation & Dexterous Manipulation

About Sarcomere Dynamics

Sarcomere Dynamics is a Canadian deep-tech company at the forefront of revolutionizing automation. By combining high-dexterity robotics with AI, we are addressing global labor shortages and increasing safety across industries. Our mission is to develop synthetic labor solutions that replicate human dexterity and adaptability, making complex automation accessible and efficient.

The Role

Sarcomere Dynamics Inc. is looking for a Software Engineer that is motivated, innovative and capable of tackling new challenging material. In this role, you will be involved in the development of integral robotic manipulation software including simulation pipelines, motion/force control, grasp synthesis, and sim-to-real transfer for robust object interaction.

What You’ll Do

• Develop control and planning stacks for dexterous manipulation: grasp planning, in-hand/bi-manual control, impedance/compliance.
• Stand up high-fidelity simulations (NVIDIA Isaac Sim, MuJoCo or similar) with accurate hand/scene models, contact, and sensing.
• Implement perception-to-action pipelines: segmentation/pose, state estimation, target region selection, grasp scoring.
• Train and evaluate policies (RL/IL/optimization-based) and integrate with classical planners (MoveIt 2, RMPflow) for hybrid control.
• Build sim-to-real tools: domain randomization, calibration, dataset generation, and performance benchmarks.
• Integrate sensors (e.g., depth cameras like Intel D455; tactile) and close the loop for robust grasps in clutter.
• Write production-grade ROS 2 nodes/services, real-time interfaces to firmware, and experiment tooling/CI.
• Manage end-to-end data collection to enable effective machine learning model training and evaluation.

Who you are

• Bachelor's degree in computer science, Robotics, Electrical Engineering, Mechanical Engineering, Physics or similar.
• Strong Python and solid C++; modern tooling (Git, Docker).
• Hands-on with at least one: Isaac Sim/Omniverse, MuJoCo, Bullet, or Gazebo/Ignition.
• Applied controls for robots: kinematics/dynamics, impedance/force control, motion planning, state estimation.
• Experience with ROS 2 (pub/sub, actions, TF, URDF/MJCF), and real-robot bring-up.
• Comfortable building datasets, evaluation scripts, and metrics; disciplined experimentation.
• Ability to work in a fast-paced environments
• Detail oriented with strong documentation abilities

Nice to Have

• Master's degree in computer science, Robotics, Electrical Engineering, Mechanical Engineering, Physics or similar.
• Grasp planning libraries, tactile sensing, or in-hand manipulation research/implementation.
• Policy learning (RL/IL) with PyTorch/JAX; sim-to-real transfer techniques.
• MoveIt 2, RMPflow, trajectory optimization (CHOMP/TrajOpt), or whole arm/hand coordination.
• CUDA/accelerated inference on Jetson-class hardware; real-time perception.
• Calibration of multi-sensor rigs; camera-hand extrinsics; latency/throughput optimization.

Compensation and benefits

• Pay range: $70k - $115k
• Paid time off
• Vision care
• Dental care
• Extended health care

Application

• Please send your resume, brief cover letter, and links to code/papers/demos to [email protected]

Job Type: Full-time

Pay: $70,000.00-$115,000.00 per year

Work Location: Hybrid remote in Pitt Meadows, BC