Apptronik
Senior Software Engineer, ML Infrastructure
Austin, TX · Senior · Full-time · Posted Aug 31, 2026
- Experience
- 5+ years of professional software engineering experience in ML platforms, data infrastructure, or related fields, OR 3+ years of direct, hands-on experience owning data and evaluation infrastructure behind models shipped to production.
- Education
- Bachelor's or Master's degree in Computer Science, Machine Learning, or a related technical field, or equivalent experience.
About the job
Builds and maintains self-serve ML infrastructure that takes robot and simulation data through curation, training, evaluation, and model deployment on humanoid robots.
What you'll do
- Build data curation, annotation, and multi-terabyte data pipelines with dataset quality, versioning, and lineage.
- Develop simulation and evaluation harnesses, model storage and promotion workflows, and packaging for robot deployment.
- Provide researcher tools for experiment tracking, training jobs, sweeps, and reproducible environments; collaborate across teams and mentor engineers.
What they ask for
- Production experience designing and shipping platform services used by other teams.
- Depth in at least one area: large-scale ML data pipelines, annotation and labeling, or evaluation and simulation; solid cloud and Python skills.
- Experience with large-scale ML data pipelines, including multimodal data, dataset versioning and lineage, and object storage.
- Strong Python and software engineering skills, plus cloud infrastructure, Kubernetes, Docker, and modern CI/CD.
Nice to have
- Robotics data formats and fleet-scale telemetry, simulation-in-the-loop evaluation, or reinforcement/imitation learning infrastructure.
- Deploying ML models to edge targets, including ONNX Runtime or TensorRT.
Tools
Python, Parquet, Arrow, lakeFS, DVC, Iceberg, S3, MinIO
AI focus
Builds infrastructure for training, evaluating, qualifying, and deploying physical AI and ML models for humanoid robots, including vision-language-action and computer-vision models.
Model-written summary · the original posting is authoritative