Samsara - Data Annotation Engineer

Job Description :

We are seeking a highly motivated and detail-oriented Data Annotation Supervisor to lead our data labeling team. This critical role ensures the accuracy, consistency, and timely delivery of high-quality annotated data, which is essential for training and validating our machine learning models. The ideal candidate will possess a strong blend of leadership skills, technical proficiency in data annotation tools and formats, and a deep understanding of data quality assurance processes.

Key Responsibilities :

Team Leadership & Management :

- Supervise and train a team of Data Annotators, ensuring adherence to project guidelines and quality standards.

- Manage daily workflow including task assignment, priority setting, and performance tracking to meet project deadlines and throughput goals.

- Conduct regular performance reviews and provide constructive feedback and coaching to annotators.

- Foster a culture of quality, efficiency, and continuous improvement within the team.

Data Quality & Process :

- Develop and maintain comprehensive annotation guidelines, providing clear examples and edge case solutions.

- Implement and manage quality assurance (QA) processes, including inter-annotator agreement (IAA) metrics and audit procedures.

- Monitor and report on key performance indicators (KPIs) such as accuracy, throughput, and consistency.

- Identify root causes of annotation errors and implement corrective actions and training to address them.

Technical & Project Coordination :

- Collaborate closely with Data Scientists, Engineers, and Product Managers to understand data requirements and model objectives.

- Select and configure appropriate annotation tools and platforms (e.g., CVAT, Labelbox, V7).

- Manage data in various formats (e.g., images, video, text, sensor data) and understand common annotation types (e.g., bounding boxes, segmentation masks, keypoints, named entity recognition).

- Proactively communicate project status, technical roadblocks, and quality issues to stakeholders.

Required Qualifications :

Experience & Education :

- Bachelor's degree in a relevant field (e.g., Computer Science, Data Science, Cognitive Science, or a related technical discipline).

- 3+ years of experience in data annotation, data labeling, or data quality assurance.

- 1+ years of experience in a supervisory or team leadership role managing an annotation or data entry team.

Technical Skills (Must-Have) :

Proficiency with Data Annotation Tools :

- Hands-on experience with at least one major annotation platform (e.g., Labelbox, Scale, CVAT, or similar internal/external tools).

Understanding of ML Data Types :

- Strong grasp of the data formats used in common Machine Learning applications (Computer Vision, NLP, etc.).

Data Formats :

- Familiarity with common data exchange formats like JSON and XML.

Quality Assurance (QA) Methodologies :

- Experience implementing and calculating quality metrics such as Inter-Annotator Agreement (IAA), precision, and recall.

Technical Skills (Nice-to-Have) :

Basic Scripting/Programming :

- Familiarity with Python or other scripting languages for data manipulation, automation of QA tasks, or light-weight data analysis.

SQL :

- Basic querying skills for retrieving and analyzing annotation logs or data from databases.

Cloud Platform Familiarity :

- Basic understanding of data storage and services on platforms like AWS, Google Cloud, or Azure.

Competencies & Attributes :

- Exceptional attention to detail and a commitment to data accuracy and consistency.

- Excellent communication and interpersonal skills for training, feedback, and cross-functional collaboration.

- Strong problem-solving ability to quickly address annotation challenges and ambiguity.

- Ability to thrive in a fast-paced, constantly evolving technical environment.