Human Data Architect, Quality

Mecka•New York, NY

1d•$130,000 - $160,000

About The Position

Mecka AI is building the data and deployment infrastructure for embodied intelligence. We collect, curate, and license the world's most useful robotics training data to leading AI labs, and we deploy real robotic systems with enterprise customers across hospitality, retail, QSR, pharmacy, logistics, and healthcare. We work with the foundation model teams shaping the next decade of robotics, and with the operators running real businesses today. Quality, trust, and execution are core to our partnerships. The Role We're hiring a Lead, Data Quality to own the quality systems, QA team, and customer-ready standard for Mecka's robotics training data. This is a data-operations leadership role, not a software QA role: you will define what good data means, measure it, improve it, and build the review systems that keep quality high as volume scales. You'll manage QA reviewers and quality analysts, partner closely with data operations, product, engineering, and customer teams, and be accountable for whether datasets are accurate, complete, consistent, and useful to frontier AI labs.

Requirements

5+ years in data quality, data operations, annotation QA, trust and safety quality, autonomy data ops, manufacturing quality, or comparable high-volume review operations.
2+ years managing reviewers, analysts, auditors, or operations leads.
Strong analytical ability with spreadsheets, dashboards, SQL, BI tools, or Python; you can diagnose quality problems with data, not anecdotes.
Experience building rubrics, SOPs, sampling plans, QA workflows, calibration programs, or customer acceptance criteria.

Nice To Haves

Built a QA system from scratch or materially improved one that was already running.
Owned customer-facing quality metrics, escalations, or dataset acceptance.
Familiarity with robotics, computer vision, video, sensor data, annotation platforms, or multimodal datasets.

Responsibilities

Define acceptance criteria for robotics datasets across labels, trajectories, video, sensor streams, metadata, task outcomes, edge cases, and customer-specific requirements.
Build rubrics, severity levels, rejection taxonomies, sampling rules, and customer acceptance gates.
Translate ambiguous customer requirements into measurable internal QA specs that operators and reviewers can execute.
Design the QA layers for each project: first-pass checks, reviewer queues, audit sampling, escalation paths, adjudication, and final release approval.
Build golden datasets, calibration tasks, inter-reviewer agreement checks, and reviewer drift monitoring.
Balance review coverage, speed, cost, and customer risk; know when to inspect, when to sample, and when to automate.
Manage, coach, and scale a team of QA reviewers, quality analysts, and review leads.
Set review SLAs, workload allocation, training standards, performance expectations, and escalation norms.
Build a quality culture where data operators own first-pass quality and QA prevents repeat defects instead of only catching them after the fact.
Own dashboards for audit pass rate, defect rate, customer rejection rate, reviewer throughput, rework, cost per approved item, and root-cause trends.
Run postmortems on quality misses and drive corrective actions across guidelines, training, tooling, and process design.
Partner with engineering to build automated validation checks for schema completeness, duplicates, time sync, metadata coverage, outliers, and model-assisted review.