Senior Software Engineer, ML Infrastructure
About The Position
Apptronik is seeking a Senior Software Engineer, ML Infrastructure to build the platform services and pipelines that handle data from collection through curation, training, and evaluation to a qualified model running on real hardware for their flagship humanoid robot, Apollo. This role involves building first-party platform services alongside open-source and commercial tooling, and it is a hands-on position on a small team whose platform is depended on daily by researchers and engineers across MLOps, Autonomy, Data Platform, and TeleOp. The goal is to turn a high volume of data from robots into shipped autonomy at a multi-terabyte scale.
Requirements
- Track record of designing and shipping production systems and services that other teams depend on daily.
- Deep hands-on experience with large-scale data pipelines for ML (multi-terabyte transformation, dataset assembly of multimodal sensor data, columnar and time-series formats like Parquet/Arrow, dataset versioning/lineage like lakeFS/DVC/Iceberg, object storage like S3/MinIO).
- Experience with ML annotation and labeling at scale (automatic annotation, human-in-the-loop workflows, tooling, quality control, throughput management).
- Experience building large-scale evaluation or simulation harnesses (many parallel jobs on GPU infrastructure, aggregated results).
- Strong Python and general software engineering ability (testing, API design, code review).
- Experience with cloud infrastructure, Kubernetes, Docker, and modern CI/CD.
Nice To Haves
- Robotics data formats and fleet-scale telemetry (MCAP, ROS, LeRobot, or equivalent).
- Simulation-in-the-loop evaluation with Isaac Sim, IsaacLab, MuJoCo, or equivalent.
- Reinforcement or imitation learning infrastructure for embodied agents (rollout workers, sim-eval harnesses).
- Deploying ML models to edge targets (ONNX Runtime, TensorRT, robot fleets).
Responsibilities
- Build the ML platform, including APIs, workers, and control planes for self-serve data and model management.
- Develop data curation and annotation workflows, including selection, filtering, automatic labeling with human-in-the-loop review, and dataset versioning.
- Implement data pipelines for routine multi-terabyte dataset operations, including transformation, assembly, coverage/quality statistics, and efficient read paths.
- Build simulation and evaluation harnesses for policy evaluation on GPU clusters, establishing benchmarks, metrics, and qualification gates for models.
- Develop the model store for versioning, metadata, evaluation results, and lineage, and manage the promotion path from trained to deployed on robot.
- Provide developer tooling for researchers, including experiment tracking, training job submission, sweeps, and reproducible container environments.
- Partner with Autonomy, Data Platform, and TeleOp on dataset and model lifecycle contracts.
- Contribute to the technical direction of ML infrastructure layers.
- Mentor engineers through code and design reviews.
Benefits
- Equal employment opportunities to all employees and applicants for employment.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
