Dyna Robotics · San Francisco
OtherResearchFullTimeposted 4w ago
Dyna Robotics builds general-purpose robots powered by a proprietary embodied AI foundation model with top-in-industry generalization and real-world performance. Already deployed with customers across multiple industries, our robots do commercial-grade work in the physical world. Our team comes from Google DeepMind, Meta, and Cruise, and we're backed by CRV, First Round, and other leading investors.
As a Research Engineer/Scientist focused on Simulation, you will build the simulated worlds that power Dyna's loco-manipulation policies. Our robot is a wheeled mobile manipulator. Mobility and manipulation are tightly coupled, not two separate problems, so simulating one without the other misses what actually matters for our policies. You'll own the pipeline end-to-end: from real-to-sim reconstruction of the facility-scale environments our robots navigate, through procedural scene generation, to sim-based training data and evaluation for policies that jointly control base motion and arm manipulation.
Real-to-Sim Reconstruction: Build a pipeline that reconstructs real, facility-scale environments into simulation (3D reconstruction, photorealistic re-rendering). That means full rooms and aisles our wheeled base navigates, not just tabletop scenes.
Facility-Scale Scene Generation: Procedurally generate navigable environments (layouts, obstacles, object placement) that stress-test loco-manipulation policies across the range of spaces our robots actually operate in.
Loco-Manipulation Sim-for-Data-Gen: Generate synthetic training data for policies that jointly reason about base positioning and arm manipulation (approach angles, reachability, obstacle-aware repositioning), and feed real-world failure modes back into new sim scenarios.
Loco-Manipulation Evaluation: Build simulation benchmarks that test the joint base+arm policy (VLA / imitation learning / diffusion models) as a whole, not manipulation in isolation on a fixed base.
Rendering for ML, Not Just Physics: Push photorealistic rendering quality specifically to close the sim-to-real gap for vision-based loco-manipulation. Rendering fidelity is a first-class deliverable, not an afterthought on top of physics accuracy.
Cross-Team Collaboration: Partner with the AI Research team on where simulated data and evaluation most accelerate loco-manipulation policy development, and with Data Ops on what's worth capturing in the real world vs. generating in sim.
MS or PhD in CS/Robotics/Graphics, or equivalent hands-on experience. We care more about demonstrated depth building simulation/rendering systems than a fixed years-of-experience bar.
Hands-on experience with simulation stacks (MuJoCo, Isaac Sim/Isaac Lab, SAPIEN, Omniverse, Blender, or similar) for robotics or graphics.
Experience with procedural scene/asset generation, domain randomization, or photorealistic rendering pipelines at scale (not academic-scale one-off scenes).
Familiarity with training/evaluating manipulation or loco-manipulation policies (imitation learning, VLA, diffusion, or RL) and how simulated data actually feeds into them.
Experience simulating mobile bases (wheeled, tracked, or holonomic) is a real plus. Most of the field's simulation talent comes from fixed-arm tabletop manipulation or legged-humanoid balance/gait work, and neither maps cleanly onto a wheeled mobile manipulator.
Prior research or engineering background in loco-manipulation or whole-body control itself, coordinating mobile-base motion with arm manipulation, independent of any simulation-specific experience.
Strong Python skills; comfort with PyTorch or JAX for anything touching model training/eval.
Senior enough to define your own research agenda and exercise independent judgment on where simulation adds the most leverage. This is closer to a founding-type ownership role than a narrow IC slot on an existing pipeline.
Experience with real-to-sim-to-real pipelines (3D reconstruction, NeRF/Gaussian splatting, differentiable rendering).
Exposure to world-model / video-prediction research (e.g. learned dynamics or latent world models) as a complement to classical physics simulation.
GPU-scale physics simulation experience (CUDA, large-batch parallel sim) rather than single-instance sim.
Background simulating wheeled/mobile manipulators specifically (e.g. Boston Dynamics Stretch, warehouse/logistics robotics) rather than legged humanoids or fixed-base arms.
Publications at CoRL, RSS, ICRA, NeurIPS, CVPR, or SIGGRAPH.
Experience leading or mentoring within a small research team.
At Dyna Robotics, we build technology for the real world, which requires a team as diverse as the environments our robots inhabit. We are an equal opportunity employer committed to technical rigor and mutual respect.
Don’t let a checklist stop you. Data shows that underrepresented groups often only apply if they meet 100% of the criteria. We value problem-solving and grit over keyword matching. If you’re passionate about robotics, no matter your discipline, we want to hear from you, even if you don't check every box.