Human Data Manager

About the role

You will ensure engineering teams get the right data they need - high-quality tasks and evaluations - by crafting and executing high-value Human Data projects. You sit at the intersection of model teams and data operations: partnering with engineering, designing projects that capture meaningful signals, and driving measurement of data impact. This is a hands-on role for people who understand training and evaluation deeply and want to influence data strategy. You will own the full lifecycle of critical data initiatives from discovery through delivery, applying rigorous validation to safeguard factuality. Work ethic and strong prioritization skills are important as you shape Grok's behavior and domain performance through targeted data management. All employees are expected to have strong communication skills to concisely and accurately share knowledge with teammates.

Key facts

What you'll do

Partner with model and engineering teams to understand needs and translate them into high-value data projects and evaluation strategies.

Own end-to-end delivery of critical data and evaluation projects that capture meaningful training signals and support rapid model development.

Leverage AI agents and existing platforms to measure data effectiveness and quantify impact including data yield, eval lift, and usage.

Maintain rigorous data integrity and truthfulness, including validation processes for factual accuracy; prioritize quality over quantity at every stage.

Research and apply techniques for data collection, annotation, generation, and multi-modal integration across text, images, and code.

Shape Grok's behavior and domain performance through targeted data work and improve annotation workflows using agents and no/low-code approaches where helpful.

Manage plans that shape model behavior via data management, optimization, and analysis, including resources, timelines, and deliverables.

Act as a liaison between engineering, technical staff, and tutoring / Human Data teams to drive alignment and knowledge sharing across functions.

Collaborate with Human Data Ops and stakeholders to scale projects, share learnings, and contribute to demand forecasting while reporting status, insights, and blockers for rapid decisions.

Define and track evaluation metrics that link data interventions to model performance improvements and inform future data strategies.

Champion data-driven decision-making by translating complex findings into clear narratives for both technical and non-technical audiences.

Optimize human-in-the-loop processes to reduce cycle time and improve signal quality without compromising factual correctness.

Identify and mitigate data bottlenecks by coordinating resources and aligning on priorities with cross-functional stakeholders.

Continuously iterate on data practices by incorporating feedback from model teams and applying lessons learned to future projects.

Requirements

Bachelor's degree, or 4+ years of relevant experience in lieu of a degree.

Experience collaborating with cross-functional teams including engineering, research, product, or annotation/operations groups.

Demonstrated experience analyzing datasets to identify trends, anomalies, quality issues, or integrity problems.

Weekend work may be required.

Travel to other SpaceXAI sites may be required.

Direct experience curating, evaluating, or improving training or evaluation datasets for large language models or other AI/ML systems.

Experience designing, supporting, or optimizing annotation workflows or data processes that prioritize factual accuracy and data integrity.

Familiarity with multimodal data such as text, images, code, or other modalities and domain-specific data including science, mathematics, programming, and recent events.

Experience with model or dataset evaluations focused on quality, truthfulness, or alignment with product goals.

Comfort using AI agents, no/low-code tools, or light scripting (e.g., Python) to prototype workflows and measurement approaches.

Experience with SQL or other data analysis tools to query, transform, and explore datasets efficiently.

Strong written and verbal communication skills to concisely and accurately share knowledge with teammates and stakeholders.

Ability to work independently and take initiative in a flat organizational structure where hands-on contributions are expected.

Commitment to maintaining rigorous standards for data quality, validation, and truthfulness in all project phases.

Willingness to scale projects, share learnings, and support demand forecasting through collaboration with Human Data Ops and stakeholders.

Nice to have

Degree in engineering, computer science, data analysis, or a related STEM discipline (Bachelor's or higher).

Direct experience curating, evaluating, or improving training or evaluation datasets for large language models or other AI/ML systems.

Experience designing, supporting, or optimizing annotation workflows or data processes that prioritize factual accuracy and data integrity.

Familiarity with multimodal data (text + images, code, or other modalities) or domain-specific data (science, mathematics, programming, recent events, etc.).

Experience with model or dataset evaluations focused on quality, truthfulness, or alignment with product goals.

Comfort using AI agents, no/low-code tools, or light scripting (e.g., Python) to prototype workflows and measurement.

Experience with SQL or other data analysis tools.

Practical notes

Weekend work may be required.

Travel to other SpaceXAI sites may be required.