ML Data Quality Engineer — CV Accuracy & Labeling

Outpost; Seattle (WA); On-site; USD 130,000 - 160,000; Full time

Job summary

Outpost is building the backbone of freight with AI-powered gate automation and CV-driven data workflows. We seek a role owner for CV accuracy across customers and yards, directing data labeling and dataset curation to improve models used in real-world gates and terminals.

You’ll own measurement, debugging misclassifications, and eventually help implement fixes and tooling for end-to-end reliability, contributing to pipeline maturation and product improvements as we scale our edge-case handling.

Qualifications

• 3+ years in data quality, ML data engineering, or applied ML.
• Production experience with computer vision or object detection.
• Proficient in Python for data analysis and pipeline tooling.
• Strong analytical rigor; ability to dig into large imagery/data volumes.
• Experience with dataset annotation tools: Roboflow, Labelbox, CVAT.

Responsibilities

• Own CV accuracy metrics reporting per customer and per identifier type.
• Investigate misclassifications, categorize root causes, and identify patterns across yards.
• Curate, label, and prioritize datasets for model retraining with ML/CV engineers.
• Build and improve the continuous learning pipeline for weekly model releases.
• Define acceptance criteria for CV accuracy per customer and track progress.
• As pipeline matures, assist in fixes and tooling for labeling and retraining.

Skills

Data quality

ML data engineering

Python

Analytical rigor

Dataset annotation tools

Tools

Roboflow

Labelbox

CVAT

Job description

Outpost is building the backbone of freight with AI-powered gate automation and CV-driven data workflows. We seek a role owner for CV accuracy across customers and yards, directing data labeling and dataset curation to improve models used in real-world gates and terminals.

You’ll own measurement, debugging misclassifications, and eventually help implement fixes and tooling for end-to-end reliability, contributing to pipeline maturation and product improvements as we scale our edge-case handling.