Sr. Staff ML Data Platform Engineer

Boston Dynamics•Waltham, MA

About The Position

We are seeking a seasoned and creative ML Data Platform Senior Staff Engineer to play a lead role scaling the data platforms that power our robots, including Spot, Stretch, and Atlas. In this role, you will shape the end-to-end data pipeline that turns raw robot logs into high-quality training data for our teams. You'll work at the intersection of robotics, data engineering, and machine learning — ensuring our models have the right data, at the right scale, with the right quality. You’ll make an impact by: Scaling the unified data platform for the robotics fleet. Grow the multi-tenant infrastructure that ingests, stores, and serves sensor data, operational telemetry, and behavioral logs from Spot, Stretch, and Atlas — supporting diverse downstream consumers including ML training, safety analysis, fleet operations, and product development. Making robot data usable at scale. Today's bottleneck isn't collecting data — it's finding and accessing the right data. You'll build the indexing, cataloging, and query layers that let engineers across the company ask questions of petabytes of heterogeneous robot data without needing to understand the underlying storage. Designing for the physical world's constraints. Robot data is messy, intermittent, multi-modal, and generated at the edge. You'll solve hard problems around schema evolution, time-series alignment across sensor modalities, graceful handling of connectivity gaps, and data integrity across on-robot, on-prem, and cloud environments. Building the platform as a product. Your users are internal engineering teams, and their adoption is your success metric. You'll define APIs, documentation, SLOs, and self-service workflows that make the platform the obvious default for any team that needs robot data — and you'll deprecate the ad-hoc alternatives that exist today. Setting the technical direction for a growing domain. Drive architecture decisions through design documents and RFCs. Build alignment across infrastructure, ML, autonomy, and operations stakeholders. Mentor engineers and set standards for a platform org that will scale alongside the fleet.

Requirements

8+ years of experience architecting and leading large-scale data platforms, ideally within autonomous vehicle, robotics, or global IoT domains.
Expertise in high-volume, low-latency data processing technologies (Spark, Kafka)
Demonstrated mastery of data warehousing solutions (e.g., BigQuery, Custom Architecture) and data models for optimizing storage and retention of massive datasets.
Strong foundation in data serialization and in-memory representation, such as Apache Arrow.
Deep expertise in 2-3 languages, including experience with performance-critical system languages (Rust, Go, or Java) and a strong foundation in systems engineering principles, memory management, and performance trade-offs.
A solid grasp of the Linux development environment and infrastructure management.
Practical experience defining and managing data lifecycles, including retention policies and regulatory compliance.

Responsibilities

Scaling the unified data platform for the robotics fleet.
Grow the multi-tenant infrastructure that ingests, stores, and serves sensor data, operational telemetry, and behavioral logs from Spot, Stretch, and Atlas — supporting diverse downstream consumers including ML training, safety analysis, fleet operations, and product development.
Making robot data usable at scale.
Build the indexing, cataloging, and query layers that let engineers across the company ask questions of petabytes of heterogeneous robot data without needing to understand the underlying storage.
Designing for the physical world's constraints.
Solve hard problems around schema evolution, time-series alignment across sensor modalities, graceful handling of connectivity gaps, and data integrity across on-robot, on-prem, and cloud environments.
Building the platform as a product.
Define APIs, documentation, SLOs, and self-service workflows that make the platform the obvious default for any team that needs robot data — and you'll deprecate the ad-hoc alternatives that exist today.
Setting the technical direction for a growing domain.
Drive architecture decisions through design documents and RFCs.
Build alignment across infrastructure, ML, autonomy, and operations stakeholders.
Mentor engineers and set standards for a platform org that will scale alongside the fleet.