Data Operations Lead — Open-Model Training (Equity)

Sierra Ventures; San Francisco (CA); On-site; USD 140,000 - 190,000; Full time

Benefits offered by this job

Top-tier compensation

Stock options

Health & wellness

Meals provided

Parental leave

Unlimited PTO

Visa sponsorship

Job summary

Reflection is seeking a Human Data Operations Manager to own the pipelines turning human expertise into better AI models. You will partner with researchers, source data from vendors, and shape data acquisition strategies to improve model usefulness and performance.

This role combines program ownership with hands-on work on task design, contributor workflows, data quality frameworks, and success metrics. Strong judgment and curiosity are essential in this fast-paced environment.

Qualifications

• 4+ years of experience delivering a portfolio of complex programs and prioritizing across changing research needs.
• Understanding of LLM training, evaluation, and prompt design, with the ability to learn unfamiliar research concepts quickly.
• Proficiency in SQL for querying datasets and building dashboards, and Python for data analysis, API integrations, and workflow automation.
• Clear written communication, strong stakeholder judgment, and a willingness to inspect individual tasks and datasets when the headline numbers do not tell the full story.
• High agency, bias towards action, and the ability to drive impact in fast-paced, dynamic, and ambiguous environments.
• Experience with annotation, expert data collection, human feedback, or AI evaluation is a plus.

Responsibilities

• Collaborate with research: Translate critical model capabilities, and researchers’ plans for developing them, into executable data acquisition strategies.
• Orchestrate providers: Assess vendors’ capabilities, forecast capacity, define task instructions, acceptance criteria and then hold vendors accountable for quality, delivery, and cost.
• Track and optimize campaigns: Build dashboards tracking throughput, turnaround time, acceptance rates, rework, and cost per accepted task. Leverage data analytics to investigate bottlenecks and guide improvements.
• Ensure alignment: Experiments and priorities evolve. You will surface vulnerabilities early and turn feedback into concrete changes to subsequent sourcing. Work with research to close the loop; tracking how our data acquisitions improve model capabilities

Skills

LLM training knowledge

SQL

Python

Stakeholder communication

Proactivity

Job description

Reflection is seeking a Human Data Operations Manager to own the pipelines turning human expertise into better AI models. You will partner with researchers, source data from vendors, and shape data acquisition strategies to improve model usefulness and performance.

This role combines program ownership with hands-on work on task design, contributor workflows, data quality frameworks, and success metrics. Strong judgment and curiosity are essential in this fast-paced environment.