Python Engineer — ML Tooling & Annotation Platform

Role overview Build and maintain the data and annotation tooling that turns large volumes of aerial imagery into dependable machine learning datasets. This owner-builder role supports the full workflow, from labeling and quality review to dataset management and reproducible training, serving both ML engineers and annotation teams.

Responsibilities - Extend annotation tools for task routing, consensus review, and model-assisted pre-labeling. - Maintain dataset infrastructure, including versioning, class taxonomies, stratified splits, and traceability from image capture to trained model. - Support reproducible training with experiment tracking, configuration management, and GPU job submission. - Build and maintain Python services, APIs, data pipelines, and command-line tools used in daily ML work. - Assess whether tools meet team needs and are being used, and simplify or remove tooling that no longer helps.

Requirements - At least four years of production Python experience developing services, APIs, or data pipelines. - Experience supporting machine learning workflows, such as annotation systems, dataset management, or training infrastructure. - Practical judgment in choosing and maintaining tools for a team's needs. - Ability to deliver clean, reliable software for ML engineers and annotation teams.