Robotics Data Quality Architect

The Consensus; New York (NY); On-site; USD 140,000 - 190,000; Full time

Job summary

Mecka AI seeks a highly skilled Human Data Architect, Quality, to define labeling rubrics, ontologies, and data schemas for robotics training data. You will shape what good data looks like and set the acceptance criteria every dataset must meet, working with researchers to align data structure with model needs.

You will own the data organization philosophy, versioning, and the path from raw data to ship-ready datasets, ensuring consistency across programs and customers.

Qualifications

• 5+ years at the intersection of ML and data in autonomy, vision, or multimodal teams.
• Hands-on experience designing taxonomies, ontologies, or labeling schemas for production model training.
• Strong data instincts; able to open a dataset in SQL/notebook and spot issues quickly.
• Comfort reading ML papers and translating model needs into data structures.

Responsibilities

• Define labeling rubrics, severity levels, rejection taxonomies and acceptance criteria for each program across data modalities.
• Translate ambiguous customer requirements into precise, measurable data specifications.
• Maintain customer-specific quality criteria and the canonical data dictionary for programs.
• Build golden datasets, reference examples, and calibration tasks to define 'correct' by demonstration.

Skills

ML data ops

Taxonomy design

Labeling schema

SQL/Python

Job description

Mecka AI seeks a highly skilled Human Data Architect, Quality, to define labeling rubrics, ontologies, and data schemas for robotics training data. You will shape what good data looks like and set the acceptance criteria every dataset must meet, working with researchers to align data structure with model needs.

You will own the data organization philosophy, versioning, and the path from raw data to ship-ready datasets, ensuring consistency across programs and customers.