Data Scientist, Chatgpt for Work

OpenAI OpenAI · AI Frontier · San Francisco, CA · Data Science

Data Scientist for ChatGPT for Work, focusing on shaping product strategy through data, defining core KPIs, building funnels, designing experiments, and translating insights into product direction for an AI-native workspace. Requires strong product sense, experimentation skills, and experience with B2B products and AI tools.

What you'd actually do

  1. Own the core KPI framework for ChatGPT for Work, spanning onboarding, activation, engagement, retention, and expansion, as well as quality/trust guardrails.
  2. Build end-to-end funnels that identify where individuals and teams succeed or get stuck, from first workspace setup through repeat usage and long-term team adoption and value creation.
  3. Define and operationalize “time-to-value” and collaboration loop metrics, and connect them to business outcomes.
  4. Design and evaluate experiments and rollouts to quantify the impact of product changes across key Work surfaces and flows.
  5. Partner with product and engineering teams to improve instrumentation, data quality, and metric definitions so decisions are fast and correct.

Skills

Required

  • 10+ years in data science / analytics in in high-velocity product environments
  • Direct experience working on B2B products (SaaS, collaboration/workspace, developer tools, or enterprise)
  • Expert SQL + strong Python
  • Strong experimentation + causal inference judgment (incl. when clean A/B tests aren’t feasible)
  • Strong product sense/taste: can turn messy signals into crisp hypotheses and roadmap direction
  • Proven ability to inspire and influence PM/Eng/Design + leadership through data storytelling
  • Autonomous operator who sets the insights/measurement agenda
  • Excellent executive communication; thrives in ambiguous, fast-moving environments

Nice to have

  • Experience with agentic and/or AI-native B2B products (agents, copilots, workflow automation, AI collaboration)
  • Experience measuring AI product quality, trust, and human-AI interaction signals
  • Familiarity with enterprise admin/security constraints and how they shape adoption
  • Experience with B2B PLG growth loops and monetization/seat expansion dynamics

What the JD emphasized

  • AI-native operator (non-negotiable)
  • super AI-pilled

Other signals

  • AI-native workspace
  • AI acts as a superassistant
  • grounding experiences in company context
  • measure, learn, and iterate on AI products