Research Engineer/Scientist, Simulation

Dyna Robotics
Redwood City
Workplace: OnsiteFull timeFunction: Research & Scientific (R&D)Skills: ["Independent judgment","Technical rigor","Problem-solving","Collaboration"]

Build and own the simulation stack that powers loco-manipulation policies for a wheeled mobile manipulator. Reconstruct facility-scale environments into photorealistic simulation, procedurally generate stress-test worlds, and generate sim-for-data training and benchmarks that evaluate joint base+arm behavior. Partner with the AI Research and Data Ops teams to close the sim-to-real gap, delivering rendering fidelity and simulation pipelines that accelerate policy development.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Dyna Robotics
Dyna Robotics
1 week ago

Research Engineer/Scientist, Simulation

âś“ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 4 hours agoStatus: Live

Job Summary

Build and own the simulation stack that powers loco-manipulation policies for a wheeled mobile manipulator. Reconstruct facility-scale environments into photorealistic simulation, procedurally generate stress-test worlds, and generate sim-for-data training and benchmarks that evaluate joint base+arm behavior. Partner with the AI Research and Data Ops teams to close the sim-to-real gap, delivering rendering fidelity and simulation pipelines that accelerate policy development.
Location: Redwood City
Workplace: Onsite
Employment Type: Full time
Job Function: Research & Scientific (R&D)

Key Responsibilities

  • •Own the real-to-sim reconstruction pipeline that converts facility-scale environments into simulation, including 3D reconstruction and photorealistic re-rendering.
  • •Procedurally generate navigable, obstacle-rich environments to stress-test loco-manipulation policies across real operating spaces.
  • •Create sim-for-data generation for policies that jointly reason about base positioning and arm manipulation, including reachability and obstacle-aware repositioning.
  • •Build simulation benchmarks to evaluate the joint base+arm policy as a whole rather than manipulation alone.
  • •Collaborate with AI Research and Data Ops to determine where simulation data and evaluation most accelerate policy development and what to capture in real-world data.

Key Requirements

  • •MS or PhD in CS/Robotics/Graphics, or equivalent hands-on experience building simulation/rendering systems.
  • •Hands-on experience with simulation stacks such as MuJoCo, Isaac Sim/Isaac Lab, SAPIEN, Omniverse, and Blender.
  • •Experience with procedural scene/asset generation, domain randomization, or photorealistic rendering pipelines at scale.
  • •Familiarity with training/evaluating manipulation or loco-manipulation policies (imitation learning, VLA, diffusion, or RL) and how simulated data feeds them.
  • •Strong Python skills, with comfort using PyTorch or JAX for model training/evaluation.
Experience:RoboticsGraphicsSimulationReinforcement learningImitation learningComputer vision
Education:
Skills:Independent judgmentTechnical rigorProblem-solvingCollaboration
Tech Stack:PythonPyTorchJAXMuJoCoIsaac SimIsaac LabSAPIENOmniverseBlender3D reconstructionNeRFGaussian splattingDifferentiable renderingCUDALoco-manipulationVLAImitation learningDiffusion modelsReinforcement learningWorld models

Company Brief

Dyna Robotics
Develops embodied-AI powered robotic arms and foundation models (DYNA-1) to automate repetitive, stationary tasks across hotels, restaurants, laundromats and retail. Focuses on affordable, deployable robots that generalize across environments and improve with on‑device learning.
Industry: Robotics
Company Size: Small (11 to 50 employees)
Growth: Growth Stage Startup
Valuation: USD 500M to 1B
Funding: Series A
Headquarters: Redwood City, United States
Founded: 2024
WebsiteLinkedInGlassdoor