About the role

We are looking for a Senior AI Data Operations Manager to own and run our end-to-end LLM model data creation pipeline from sourcing and managing vendors, to coordinating with engineering teams to evaluate model performance to closing the loop by creating targeted data for edge cases ensuring the data our models train on meets the highest quality standards. This role will manage the end-to-end AI data pipeline across 10+ AI features spanning multiple products on our platform

This is a cross-functional operations role that sits at the intersection of ML, product, and operations management. You will be the connective tissue between our machine learning engineers, internal annotators, and external vendors, translating technical requirements into actionable data creation workflows and ensuring every batch of data that reaches our models is accurate, well-annotated, and fit for purpose. Critically, you will also drive automation across the pipeline, designing and deploying AI agents that eliminate manual bottlenecks and make the entire operation smarter over time.

What you will do

• Own the full AI feature development lifecycle, from sourcing data to quality management and relationship building at scale.
• Design and deploy agentic workflows that automate multi-step processes including vendor briefing, annotation verification, and batch tracking.
• Translate model evaluation signals (precision gaps, recall failures, edge case blind spots) into targeted data creation cycles.
• Partner with EMs, MLEs and PMs to stay ahead of model data needs, and close the loop between evaluation findings and pipeline improvements.
• Maintain clear, living documentation across data batches, annotation guidelines, and vendor SOPs.

Who you are

• 5+ years of technical program management/operations management with 3+ years in AI/ML data operations, LLM pipeline management, and vendor management in a tech environment.
• AI-fluent: you use AI tools naturally, think in terms of what can be automated, and can hold credible conversations with both technical and non-technical stakeholders about model behavior, prompt design, and agent architecture.
• Hands-on experience building or working with AI agents. Organized and detail-oriented, comfortable managing multiple vendors, deadlines, and pipelines simultaneously.
• Familiar with LLM evaluation basics (precision, recall, edge cases) and can engage meaningfully with engineers on data quality.

Even better if you have

• Prior experience working with NLP or AI training data pipelines.
• Familiarity with annotation tools (e.g. CVAT, Label Studio, Scale AI, Surge AI, or similar).
• Experience working in a startup or high-growth environment where processes are still being built.

You will thrive in this role if you

• Get frustrated when things are manual that don't need to be, and you do something about it.
• Can translate messy technical requirements into clear, actionable briefs without losing anything in translation.
• Treat data quality as a product, not a checkbox.
• Are comfortable operating in the gap between engineering and operations, and you actually enjoy being there.