Data Engine & Annotation Systems Engineer (Annotation Systems Engineer)

Simbe; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Comprehensive health plans

Culture of learning

Flexible time off

Job summary

Simbe is seeking a Data Engine & Annotation Systems Engineer to own the systems, workflows, and tooling powering high-quality training data for our computer vision models.

You will build the data engine behind Simbe’s AI platform, implementing model-assisted labeling, data quality checks, annotation guidelines, error mining, dataset versioning, active learning, and evaluation workflows that improve model performance and accelerate customer value.

Qualifications

• 3+ years of experience in software engineering, data tooling, ML data operations, annotation systems, data QA, or related technical work.
• Comfort with Bash, Linux, Git and production debugging workflows.
• Strong communication skills with ability to create clear annotation guidelines and documentation.
• Experience with web frontend or full stack development for internal tools.
• Experience with CVAT, FiftyOne, Labelbox, Scale, Roboflow, Supervisely, or similar tools.
• Experience with annotation/data quality workflows in CV/AI projects.
• Experience with object detection, segmentation, OCR, or related CV tasks.

Responsibilities

• Own annotation systems and workflows, including setup, guidelines, quality control, edge case handling, and throughput monitoring.
• Build data engine tooling using Python, web, and automation to request annotations, review results, clean data, export datasets, and evaluate model performance.
• Improve data quality by designing checks for label consistency, bounding box sizes, missing annotations, duplicates, class imbalance, and other quality issues.
• Integrate model-assisted workflows such as auto annotation, pre-labeling, active learning, and hard case mining to boost efficiency.
• Support dataset versioning and evaluation in collaboration with CV engineers for releases and benchmarks.
• Coordinate across annotation teams, CV, product, customer success, and data teams to align systems with customer priorities.
• Monitor performance metrics like quality, turnaround time, capacity, cost, rework, and model impact.

Skills

Python

Strong communication

CV/annotation domain

Active learning

Tools

Bash

Linux

Git

CVAT

FiftyOne

Labelbox

Scale

Roboflow

Supervisely

Job description

• Simbe is looking for a Data Engine & Annotation Systems Engineer to own the systems, workflows, and tooling that power high quality training data for our computer vision models
• This role goes beyond annotation coordination. You will help build the data engine behind Simbe’s AI platform: model assisted labeling, data quality checks, annotation guidelines, error mining, dataset versioning, active learning, and evaluation workflows that improve model performance and accelerate customer value
• You will help create the feedback loop that makes Simbe’s AI systems better every week
• You will improve the quality, speed, and reliability of data used to train and evaluate production models
• You will work across human annotation, automation, model outputs, QA, and customer impact
• Own annotation systems and workflows. Oversee and improve image and video annotation workflows, including task setup, guideline creation, quality control, edge case handling, and throughput monitoring
• Build data engine tooling. Develop Python, web, and automation tools that make it easier to request annotations, review results, clean data, export datasets, and evaluate model performance
• Improve data quality. Design checks that identify inconsistent labels, oversized or undersized boxes, missing annotations, duplicate data, class imbalance, and other issues that can degrade model performance
• Integrate model assisted workflows. Evaluate and integrate auto annotation, pre labeling, active learning, hard case mining, and model in the loop review systems to improve annotation efficiency
• Support dataset versioning and evaluation. Partner with CV engineers to maintain reliable datasets, evaluation splits, benchmark views, and release readiness workflows
• Coordinate across teams. Work with annotation teams, Computer Vision, Product, Customer Success, and Data teams to make sure annotation systems support current customer priorities and future product needs
• Measure and improve performance. Track annotation quality, turnaround time, capacity, cost, rework rates, and model impact to improve the operating system for data creation

Benefits

• Comprehensive health plans
• Culture of learning
• Flexible time off

High attention to detail and a strong understanding of how data quality affects model quality3+ years of experience in software engineering, data tooling, ML data operations, annotation systems, data QA, or related technical workComfort with Bash, Linux, Git, and production debugging workflowsStrong communication skills and ability to create clear annotation guidelines, documentation, and process improvementsAbility to coordinate across technical and operational teams while still writing code and improving systemsExperience with web frontend or full stack development for internal toolsStrong Python skills, including experience building scripts, data workflows, APIs, or internal toolsExperience with CVAT, FiftyOne, Labelbox, Scale, Roboflow, Supervisely, or similar annotation and dataset toolsExperience with image, video, robotics, retail, autonomous vehicle, industrial inspection, or sensor data annotationExperience with active learning, auto labeling, synthetic data, evaluation dashboards, or dataset versioningExperience with object detection, segmentation, OCR, barcode localization, or other computer vision workflowsExperience managing external annotation vendors or distributed annotation teams