Senior Staff Data Scientist, Robotics

Agility’s commercially deployed humanoids operate alongside teams in warehouses, manufacturing facilities, and distribution centers—tackling physically demanding and repetitive tasks while enabling workers to focus on higher-value work. With industry-leading safety standards and years of proven deployment data, we're pioneering a new era of automation that enhances human potential.

About The Role

Agility Robotics builds Digit, a humanoid robot deployed into real warehouses and factories under a Robots-as-a-Service (RaaS) model. Every hour Digit operates generates telemetry, logs, sensor streams, and maintenance events — and turning that into reliability, unit economics, and product decisions is the job.

As a Senior Staff Data Scientist, you'll define how we use data to build better robots and make better decisions. This is a high-impact, high-visibility role where you'll set the technical direction for analytics and modeling while helping build a truly data-driven engineering organization.

In this newly created role, you'll transform massive volumes of complex robot data into the insights and models that drive decisions across hardware, software, manufacturing, and operations. You'll create the foundation for how we measure success, improve robot performance at scale, and prioritize what to build next — accelerating how we design, deploy, and continuously improve our autonomous systems.

About The Work

Predictive maintenance & hardware reliability . Build models that predict MTBF and remaining useful life for specific components (actuators, cameras, compute, power systems). Partner with hardware engineering to model wear-and-tear under varying duty cycles, payloads, and environmental conditions, and turn those models into maintenance schedules and design feedback.

Fleet performance & RaaS unit economics . Analyze telemetry and logs to find which software versions, site conditions, or usage patterns correlate with the highest failure and intervention rates. Work with Product to define the "golden signals" of a RaaS deployment and stand up the dashboards behind each. Quantify the cost of human intervention (teleop, on-site support, manual recovery) and its drivers.

Root-cause & anomaly detection tooling . Build detection and RCA tooling that surfaces anomalies in fleet behavior early and helps engineers get from symptom to cause faster.

Manufacturing quality & feedback loops . Join end-of-line test data with field performance to find which manufacturing signals predict early field failures, and close the loop back to the factory to catch defects before they ship

Beyond the original charter, you may also help shape:

Experimentation & fleet A/B — a framework for safely rolling out software/firmware changes across a physical fleet and measuring impact on performance, reliability, and intervention cost.

Data quality & instrumentation strategy — partnering with embedded/software teams to define what gets logged and at what fidelity, so the data needed for these models exists in the first place.

Demand/capacity & deployment economics — models that inform fleet sizing, spares/inventory, and the economics of new site rollouts.

Safety statistics .

About You

10+ years applying data science / statistical modeling to real-world problems, with a track record of owning ambiguous, high-impact problems end to end.

MS or PhD in Statistics, Computer Science, Operations Research, Engineering, Physics, or a related quantitative field; or equivalent professional experience. PhD and coursework/experience in reliability engineering, survival analysis, or applied statistics preferred.

Prior work in robotics, autonomous systems, hardware, IoT/connected devices, industrial, or manufacturing settings.

Deep expertise in some combination of: reliability/survival analysis, time-series and anomaly detection, predictive maintenance, and causal/observational inference.

Strong software fundamentals — production-quality …