SpaceX
Full-Stack Data Scientist, Hardware Reliability (Starlink)
About this role
SpaceX seeks a Full-Stack Data Scientist to build data pipelines, machine learning models, and production systems that improve Starlink hardware reliability. You'll work with telemetry, manufacturing, and support data to deploy predictive models and LLM-based tools that drive replacement and support decisions across satellites, dishes, and customer equipment.
What you'll do
- Build and maintain data pipelines and infrastructure to increase Starlink hardware reliability
- Design feature engineering, labeling, and scoring systems for production and field models
- Train, validate, and deploy hardware failure prediction models in production environments
- Operate production models including monitoring, versioning, and iteration based on live results
- Build internal analytics tools and APIs for hardware, production, quality, and supply chain teams
- Collaborate across engineering and operations to translate analyses into design and process improvements
What they're looking for
- Python and SQL
- Production data pipeline development
- Machine learning model deployment and monitoring
- Relational databases and data orchestration
- Large language models (RAG, tool calling, agentic workflows)
- Statistical analysis and reliability methods (survival analysis, calibration)
- Software engineering fundamentals (data structures, algorithms, system design)
- Unstructured data extraction (tickets, chat, images)
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
SpaceX
SpaceX develops advanced spacecraft and satellite systems, including the Starshield government satellite constellation and Starfall re-entry cargo capsule for global delivery. The company is hiring engineers in avionics integration, software test automation, mechanical design, and hardware reliability to validate flight-critical systems and ensure mission success.
- Website
- spacex.com
Likely interview questions
- Walk us through a production ML system you've built—how did you handle model versioning, monitoring, and retraining?
- Describe your experience working with unstructured data like support tickets or images. How did you extract signal and integrate it with structured data?