Company Overview / Who We Are

Purpose is an AI-powered personal growth mentoring app built to help people make better decisions, gain clarity, and live more aligned lives. Co-founded by Mark Manson, three-time New York Times bestselling author and repeat AI founder Raj Singh, Purpose brings thoughtful mentorship to people at a scale that was never possible before. Since launching in Q4 2025, Purpose has grown quickly, growing to tens of thousands of new users within a month and supporting a global community. With strong foundations in place, we’re entering our next phase of growth and building a world-class team to help shape what comes next.

Most AI coaching apps are expensive journaling. You talk, it validates, you feel heard, nothing changes. We are building something different. Purpose uses structured coaching frameworks, validated psychometric assessments, and a quest-based scaffolding system grounded in behavior-change science. We measure whether someone is getting better, not whether they opened the app today.

This is not another chatbot. This is a team working to get it right, and to publish what it learns.

About the Role

We treat everyone here as a product scientist. You're not implementing specs; you're responsible for whether this product actually helps the people who use it.

We're hiring a Staff Engineer to own the AI science and data systems behind Purpose's coach: how we evaluate it, how it remembers, and how we measure whether people are changing for the better. You'll report to the Chief Science and Technology Officer and partner closely with our Product Manager, Behavioral Science.

The questions in this role are still open across the field. How do you measure whether an AI mentor helps someone change, not just whether they feel heard? How does a coach remember what matters across months of conversation, and know what not to bring up? How do you personalize to one person rather than the population average, without compromising the validity of what you measure? These questions are our roadmap. You'll work on them with real users, real data, and support to publish what you learn.

Internally, we use Member of Technical Staff titles. It's a flat organization where the best ideas win, not the loudest voice or the highest title. You'll have real autonomy to set priorities within your area, share what you find, and protect the important work that isn't urgent. This role is dedicated to AI science and data systems; it is not a product analytics or reporting role. Expect roughly 60% of your time on AI science and experiments and 40% on the data foundation those experiments depend on, with the balance shifting as needs change.

Location: Remote

Time Commitment: Full-time

What You'll Do

• Evaluation and quality
  • Build and run the systems that tell us whether the coach is good and safe: LLM-as-judge evaluation, human annotation workflows, evaluation datasets, and the quality gates that prompt, model, and cost changes must pass before they ship.
• Memory
  • Own how the coach remembers. Improve retrieval quality, design memory that grows with a person over months, and build the evaluations and safeguards that keep it accurate and appropriate.
• Measurement infrastructure
  • Own the data layer that gives every user insight a consistent name, type, and set of permissions, so chat, quests, notifications, and research can all build on the same trusted foundation. This is the basis for personalization and for the outcome claims we can credibly make.
• Experiments and model decisions
  • Design and run the experiments behind our AI decisions: model and provider comparisons, prompt optimization, and A/B tests on coaching quality. Bring the statistical rigor that keeps those decisions honest.
• Data privacy and security
  • People share deeply personal things with Purpose. You'll own the architecture and practices that keep that data safe and worthy of their trust.