Senior ML Platform Engineer, Robotics
Andromeda Robotics
·Yesterday
- Salary
- $150k – $220k/yr
- Location
- San Francisco, California
- Workplace
- Onsite
- Type
- Full-time
- Department
- Engineering - ML&AI & Data
- Seniority
- Senior
- Source
- Workable
Description
The Bigger Picture
At Andromeda, we're building the social intelligence layer for humanoid robots - technology that
allows robots to understand people, build relationships and become a meaningful part of everyday life.
Our first robot, Abi, is already doing this in aged care and healthcare, where she brings conversation, companionship and engagement to people every day.
Now we're scaling from early deployments into a platform that can operate across thousands of real-world environments. We're looking for individuals who want to tackle hard problems across robotics and AI, and help define what it means for robots to live and interact alongside people.
The Role
You'll build the shared workflows behind our ML development: reproducible experiments, trustworthy evaluations, dependable training data, reliable training infrastructure, and models that integrate with the robot software we already run. You'll be the champion for ML developer experience, making the best-practice workflow the fastest and easiest path for the team.
This is a hands-on engineering role. You'll write the first implementation, help teams use it, and improve it against real workloads. Some days that means making a training job recover cleanly. Others, it means finding why a promising model behaves differently on Abi. This is a senior individual contributor role. You will lead by influence and set technical direction for ML tooling, not manage a team.
You'll partner with our existing DevOps, cloud, and robot software platform engineers and extend it for ML workflows.. Our stack includes, AWS, Nix, ROS 2, and NVIDIA Jetson, third party and custom tooling. What we need is strong MLOps judgment, the drive to learn, and an obsession with making your teammates' workflows effortless.
This role is on-site five days a week in SF
What you'll own
- Repeatable experimentation. Establish practical patterns for experiment repositories, configuration, code review, and automated checks. Build reusable libraries, scripts, and templates that let engineers start quickly without leaving behind work only its author can run.
- Data infrastructure. Build the multimodal data lake and pipelines that clean, annotate, and serve data for model training. Add quality checks and lineage so every dataset traces back to the robot data it came from.
- Experimentation and evaluations. Make it cheap to test new ideas against production baselines and challenger models, and clear when a candidate genuinely wins. Define the path a model follows: candidate, qualified against agreed criteria, approved, then deployed to the robot.
- Develop patterns for robot integration testing. Test the complete model path inside the existing robot codebase: inputs, preprocessing, inference, interfaces, and downstream behavior. Work with engineers on CI, simulation, and hardware-in-the-loop testing that catch failures that software only testing will miss
- Cloud and training infrastructure. Own the ML-specific workflows for provisioning compute, scheduling and managing jobsWork with cloud and DevOps engineers on the underlying infrastructure, access, reliability, and cost.
- Reproducible environments and deployable models. Build reliable packages and dependency configurations. Learn and apply Nix packaging, the Nix store, and binary caches. Contribute to existing robot build and release workflows so models reach Jetson with compatible dependencies, tested runtime behavior, and a rollback path. Support staged rollouts or shadow deployments, and monitor how deployed models behave on robots in the field.
- Product-linked reporting. Build dashboards that connect model and dataset versions, evaluation results, and deployed performance to agreed product metrics. Help teams distinguish benchmark gains from improvements in Abi's behavior, reliability, and user experience.
- Adoption across the engineering team. Work alongside engineers to understand bottlenecks, advise on technical choices, and make shared tools useful in practice. Provide clear examples, documentation, and handoffs.