Data Scientist, ChatGPT for Work at Openai

Openai

San Francisco·19h ago

Mid LevelFull-timePythonRustRailsAWSData ScienceAIData

Salary Range

$80,750 - $109,250 /year

Estimated

This salary is estimated based on similar roles. The actual salary may vary.

About the Role

We are looking for a Data Scientist to join OpenAI's ChatGPT for Work team. You'll shape product strategy through data analysis, uncovering user problems and forming hypotheses to improve team and business outcomes. You'll be responsible for defining success metrics, identifying bottlenecks, and turning insights into product direction.

Key Responsibilities

→Own the core KPI framework for ChatGPT for Work, covering onboarding, activation, engagement, retention, and expansion
→Build end-to-end funnels to identify where individuals and teams succeed or get stuck
→Define and operationalize metrics like "time-to-value" and collaboration loop metrics
→Design and evaluate experiments to quantify the impact of product changes
→Partner with product and engineering teams to improve data quality and metric definitions
→Translate complex analysis into clear insights that shape product strategy
→Help establish data science standards for measuring human-AI collaboration
→Share learnings with other data scientists across the company

Requirements

✓10+ years in data science/analytics in high-velocity product environments
✓Direct experience working on B2B products (SaaS, collaboration/workspace, developer tools, or enterprise)
✓Expert SQL and strong Python skills
✓Strong experimentation and causal inference judgment
✓Strong product sense: ability to turn messy signals into crisp hypotheses
✓Proven ability to influence PM/Eng/Design and leadership through data storytelling
✓Autonomous operator who sets the insights/measurement agenda
✓Excellent executive communication; thrives in ambiguous, fast-moving environments
✓AI-native operator: first to adopt new AI tools and integrate them into workflows
✓Why it matters
✓This role is critical because the way we measure, learn, and iterate will determine whether AI becomes a genuinely trusted superassistant and coworker for everyday knowledge work, at scale.