Data & Infrastructure Engineer

The role

You will own the path from robot site to a searchable, trustworthy dataset that researchers train on and customers buy. You will also run the infrastructure that keeps a growing, geographically spread fleet updated, monitored, and healthy.

What you will do

• Own data storage and lifecycle: ingest from sites, object storage layout, retention, tiering, and cost.
• Build the catalog: episode metadata, task labels, operator and robot info, quality flags, and search.
• Build automated quality checks: dropped frames, sync drift, missing streams, failed or truncated episodes.
• Build dataset export and delivery: convert raw episodes into training and customer formats, with versioning and provenance.
• Handle data governance: access control, customer and site data agreements, privacy, and audit trails for data we sell.
• Run fleet infrastructure: provisioning edge machines at sites, remote software deploys and rollbacks, secrets, VPN/connectivity, monitoring, and alerting.
• Keep cloud costs and bandwidth under control as data volume grows.
• Build dashboards to monitor data collection status and progress.

Required qualifications

• 3-4 years as a data, platform, or infrastructure engineer owning production systems end to end.
• Experience with large unstructured data (video, sensor logs, images) in object storage, not only tabular warehouses.
• Strong Python and SQL.
• Comfortable with infrastructure-as-code, containers, and CI/CD.
• Experience building pipelines with real failure handling: retries, idempotency, backfills, and monitoring.

Nice to have

• Experience with robotics or autonomous-vehicle data, or ML training data pipelines.
• Edge fleet management: deploying to many remote Linux devices on unreliable networks.
• Experience packaging and delivering datasets to external customers.
• Familiarity with data privacy rules across multiple countries.

Before you apply

• The posting says you can apply if you do not match every requirement but believe you can do the job.

About the company

Mecka AI builds the data infrastructure layer for robotics and embodied AI. It designs and operates global systems for data capture, labeling, and hardware-enabled workflows used by AI labs and robotics companies.