ML Systems Engineer (Platform & Biometrics Data Infrastructure)

Eight Sleep•San Francisco, CA

About The Position

Join the Sleep Fitness Movement At Eight Sleep, we’re on a mission to fuel human potential through optimal sleep. As the world’s first sleep fitness company, we’re redefining what it means to be well-rested and building the most advanced hardware, software, and AI technology to make it possible. Our products power peak mental, physical, and emotional performance by transforming every night of sleep into a personalized, data-driven recovery experience. We are trusted by high performers, professional athletes, and health-conscious consumers in over 30 countries worldwide. Recognized as one of Fast Company's Most Innovative Companies in 2019, 2022, and 2023, and twice named to TIME's “Best Inventions of the Year.” We operate like a high-performance team: fast, focused, and motivated by impact. We don’t just ship; we iterate, refine, and obsess over the details that help our members sleep better and wake up stronger. Every role at Eight Sleep is a chance to create cutting-edge technology, collaborate with world-class talent, and help shape a future where sleep isn’t passive - it’s a powerful tool for living better. If you’re tired of the ordinary and driven to build at the edge of what’s possible, this is your moment. Join us and lead the movement that’s transforming how the world sleeps and what we’re all capable of when we wake up. High Standards. No Apologies We operate with intensity because our mission demands it. At Eight Sleep, we bring the same mindset as the world’s top performers: focused, relentless, and always pushing to be in the top 1% of our craft. Think Kobe Bryant’s mamba mentality, applied to bold ideas, next-gen tech, and flawless execution. This isn’t a 9-to-5. We’re a team that puts in the extra effort, not because it’s required, but because we care about the impact of our work. We’re here to build fast, push limits, and deliver without compromise. If you thrive under pressure and want to do the most meaningful work of your career, you’ll feel right at home. If you’re looking for something easier – this isn’t it. The role We’re hiring an ML Systems Engineer / Data Engineer to build the platform foundations that power our ML lifecycle - from data ingestion and curation to training, evaluation, deployment, and monitoring. This is a highly leveraged role: your work will accelerate every model, every experiment, and every product launch across AI/ML. You’ll partner with ML engineers, data science, firmware, and backend teams to create scalable, reliable systems for time-series and biometrics-grade data.

Requirements

2+ years building data/ML infrastructure in production (data engineering, ML platform, or MLOps).
Strong Python engineering and SQL fluency; proven ability to write clean, maintainable, high-performance code.
Experience with distributed processing (e.g., Spark, Flink, Ray) and workflow orchestration (e.g., Airflow or similar).
Experience with cloud infrastructure and data systems (e.g., AWS/GCP; object storage; streaming systems; warehouses/lakehouses).
Practical understanding of ML development workflows (training/eval/inference), and how platform decisions affect model velocity and reliability.
Strong debugging skills in Linux environments and comfort operating systems that run 24/7.

Nice To Haves

Experience with time-series, sensor, or biometrics data (wearables/IoT) and handling irregular sampling, missingness, and calibration.
Experience building feature stores, online/offline parity systems, and low-latency inference pipelines.
Experience with containerization and deployment (Docker/Kubernetes), and ML CI/CD practices.
Exposure to regulated or research environments (clinical studies, validation processes, privacy/security constraints).

Responsibilities

Build and operate high-throughput pipelines for sensor and event data (batch + streaming), with strong guarantees on quality, lineage, and reliability.
Create scalable dataset curation and labeling workflows: sampling, slice definitions, weak supervision, gold-set management, and evaluation set integrity.
Develop ML platform components: feature pipelines, training orchestration, model registry, reproducible experiment tracking, and automated evaluation.
Implement monitoring and observability for production ML systems: data drift, performance regression, alerting, and automated failure detection.
Standardize schemas and interfaces across studies and product telemetry to enable reusable, consistent analytics and model development.
Collaborate cross-functionally to support new studies and launches, ensuring data architecture meets evolving research and product needs.