Robotics Infra DevOps Engineer

Socket.dev; Boston (MA); On-site; USD 120,000 - 180,000; Full time

Job summary

FieldAI is seeking a DevOps Engineer to own the humanoid research infrastructure from CI/CD to data pipelines and security. You will build and own the stack, work with Field AI’s platform, cloud, and data-processing teams, and drive reliability for ongoing humanoid robotics work.

You will join a mission-driven team based in the Boston/Cambridge area, focusing on scalable, testable pipelines, reproducible environments, and pragmatic velocity in a research setting.

Qualifications

• 4+ years in DevOps, platform, infrastructure, or SRE roles.
• AWS in production: ECR, S3, IAM, VPC, EKS, Batch, ECS/Fargate.
• Infrastructure as code (Terraform, CDK, Pulumi, or equivalent) with a review-and-version discipline.
• CI/CD design and operation at scale — GitHub Actions strongly preferred, including self‑hosted runners.
• Kubernetes in production, including workload scheduling and resource governance.
• Container tooling and build optimization: Docker, BuildKit or daemonless alternatives (Buildah, Kaniko), multi‑arch builds, remote caching.
• Workflow orchestration — Apache Airflow or equivalent.
• Python, plus comfort in Bash and reading C++
• Linux systems administration and networking fundamentals
• Observability: logging, metrics, tracing, and alerting you actually built
• Comfortable being the only infra person in the room; bias toward reuse; strong docs; pragmatic about research velocity

Responsibilities

• Stand up CI/CD for the humanoid monorepo: containerize, push to ECR, run unit tests, build, and gate on simulation system tests before promotion.
• Work with the platform team's self-hosted GitHub Actions runners (ARM, AMD, CUDA, Jetson-class targets).
• Cut build times through change detection and remote caching.
• Build test infrastructure for sim, Isaac Sim, or real hardware via ROS2 messages or REST API.
• Establish per-automation integration tests for shared-library and output-format changes.
• Stand up and own FAIRI's Airflow stack for humanoid data processing.
• Build the ingest path from robot to dataset: rosbag/MCAP capture, episode segmentation, conversion, delivery to training.
• Implement data lifecycle guardrails—filtering, deletion reviews, retention.
• Build and operate dataset and mission registry for attribution to subject/session/robot/purpose.
• Support MoCapDB: ECS Fargate, AWS Batch retargeting workers, RDS Postgres, S3, with FieldAI Auth.
• Own FAIRI's AWS footprint as code: ECR, S3, IAM roles and cross-account trust policies, VPC and networking, Kubernetes/EKS workloads.
• Close gaps where infrastructure is not yet in code, bring permissions changes under review.
• Own compliance posture for research tooling—SOC 2 constraints on SaaS, experiment tracking, data-sharing controls.
• Eliminate person-owned infrastructure: documented owners, runbooks, and access paths.
• Manage secrets, VPN/Tailscale access paths, and hardware-in-the-loop connectivity to robots on the floor.
• Write and maintain runbooks, onboarding guides, and architecture docs.
• Be the interface between FAIRI and Field AI's platform, cloud, and data-processing teams; negotiate reuse vs own.
• Support researchers and systems engineers during demos and live troubleshooting; ensure reproducibility with versioned configs

Skills

DevOps

AWS

CI/CD

Kubernetes

Terraform

Python

GitHub Actions

Docker

Tools

BuildKit

ECR

S3

EKS

RDS

Ansible

Job description

FieldAI is seeking a DevOps Engineer to own the humanoid research infrastructure from CI/CD to data pipelines and security. You will build and own the stack, work with Field AI’s platform, cloud, and data-processing teams, and drive reliability for ongoing humanoid robotics work.

You will join a mission-driven team based in the Boston/Cambridge area, focusing on scalable, testable pipelines, reproducible environments, and pragmatic velocity in a research setting.