2 months ago
Responsibilities
- Build APIs, workers, and control planes for self-serve ML data and model workflows.
- Develop data curation, annotation, quality-control, versioning, and lineage workflows for synchronized robot and simulation data.
- Build multi-terabyte data transformation, dataset assembly, statistics, and GPU-efficient read pipelines.
- Create GPU-based simulation rollout harnesses, benchmarks, metrics, and automated model qualification gates.
- Develop model stores and promotion workflows from trained models through qualification to robot deployment, including ONNX and TensorRT packaging.
- Provide experiment tracking, training job submission, sweep, and reproducible container tooling for researchers.
- Partner with Autonomy, Data Platform, and TeleOp on lifecycle contracts and technical direction.
- Mentor engineers through code and design reviews.
Requirements
- 5+ years of professional software engineering experience in ML platforms, data infrastructure, or related fields, or 3+ years of direct hands-on experience owning production model data and evaluation infrastructure.
- Production-scale experience designing and shipping platform services used daily by other teams.
- Hands-on experience with large-scale ML data pipelines, multimodal sensor data, dataset versioning and lineage, and object storage.
- Experience with large-scale ML annotation and labeling workflows combining automatic annotation with human review.
- Experience building evaluation or simulation harnesses with parallel jobs on GPU infrastructure.
- Strong Python and general software engineering skills, including testing, API design, and code review.
- Experience with cloud infrastructure, Kubernetes, Docker, and modern CI/CD.
- Bachelor’s or Master’s degree in Computer Science, Machine Learning, or a related technical field, or equivalent experience.
- Bonus: robotics data formats and fleet-scale telemetry such as MCAP, ROS, or LeRobot.
- Bonus: simulation-in-the-loop evaluation using Isaac Sim, IsaacLab, or MuJoCo.
- Bonus: reinforcement or imitation learning infrastructure for embodied agents.
- Bonus: deploying ML models to edge targets using ONNX Runtime or TensorRT.
Benefits
- Direct-hire position.
- Physical requirements include prolonged desk and computer work, occasional lifting of up to 15 pounds, and the ability to read screens and communicate by hearing and speech.
Tech Stack
Categories
Data EngineeringML Engineering
About Apptronik
Apptronik designs and manufactures AI-enabled humanoid robots, led by its Apollo platform, to automate tasks alongside people in manufacturing and logistics, with potential use in healthcare and homes. The privately held company is headquartered in Austin, Texas, and grew out of the University of Texas at Austin’s Human Centered Robotics Lab; its team previously worked on NASA’s Valkyrie robot. Apptronik’s business model centers on selling robots and the software and services needed to deploy them in industrial environments.