Data Operations Engineer

Labelbox • San Francisco, CA

3d • Hybrid

About The Position

We are seeking a skilled and detail-oriented Data Operations Engineer to support our data annotation and data quality assurance processes. In this role, you will play a critical part in optimizing, maintaining, and scaling our data labeling workflows, primarily using Labelbox. You will ensure that labelers are able to efficiently and accurately generate human-labeled data by building tools, using LLM models, automating common project management tasks, and troubleshooting complex issues within the production pipeline. Your ability to script in Python and apply engineering problem-solving principles to data operations will be key to improving both efficiency and quality across our projects.

Requirements

• 3+ years of working experience in a technical role, interfacing with technical and non-technical teams, and writing Python scripts for data processing.
• 2+ years of experience using LLMs in prompting frameworks (e.g. LLM-as-a-judge).
• Some experience with machine learning models in scripts or data pipelines.
• Bachelor’s Degree in Engineering, Computer Science, or a technical field.
• Practical experience using LLMs or traditional models to assist annotation QA or generate/transform data.
• Proficiency in Python scripting and experience with automation of operational tasks.
• Experience with Labelbox or similar data annotation platforms.
• Strong analytical and problem-solving skills with a demonstrated ability to optimize processes.
• Experience with data pipelines, data analysis, and data workflow management.
• Familiarity with cloud platforms such as AWS, GCP, or Azure.
• English fluency.
• Knowledge of Statistical Analysis techniques to uncover bad patterns in human-labeled data.

Nice To Haves

• Experience with Labelbox or similar data annotation platforms.
• Prior experience in a production or process engineering role, especially in data operations or similar environments.
• Understanding of project management methodologies and the ability to work collaboratively across teams.

Responsibilities

• Build, deploy, and maintain Python automation scripts and other tools to streamline the data annotation process, automate repetitive tasks, and reduce manual effort.
• Identify bottlenecks in the data labeling pipeline and implement solutions to enhance throughput, accuracy, and scalability of labeling operations.
• Work closely with the Project Management team to ensure that data labeling meets accuracy standards and troubleshoot any issues related to data quality.
• Plan quality assurance workflows to use GenAI and open-source models to find data anomalies.
• Set up monitoring tools to track the performance of data annotation operations, reporting key metrics and areas for improvement to leadership.
• Integrate and manage third-party api tools with Labelbox, ensuring seamless operation and data flow across platforms.
• Ability to build and maintain internal tools with retool and similar tools.
• Provide ongoing technical support to the project managers and labelers, assisting with technical challenges in Labelbox and associated tools.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Number of Employees: 101-250 employees