Responsibilities

• Lead the planning, outlining, and generation of the depth associated with each learning node, from hand-selected golden examples to AI generation at scale.
• Design clear, approachable content that meet high quality bars for clarity, difficulty ramp, voice, and bingeability 1 then hold the AI to those bars.
• Design and monitor evals that keep 3almost right4 from shipping; monitor drift and regressions and course-correct quickly.
• Use data and learner feedback to prioritize coverage, tune difficulty, and address confusion clusters; iterate quickly to improve outcomes.
• Prototype pragmatic tooling that speeds up high-quality depth creation and bulk edits to prove value before Engineering hardens them.
• Document playbooks so other producers can run these flows and hit the same quality bar.
• Partner with Product, Design, and Engineering to influence the representations and interfaces that make interactive content reliable and LLM-operable.
• Frequently ask: How does this impact our learners?

Requirements

• You have strong pedagogical taste in STEM education and can articulate what 3good4 looks like.
• Are AI-literate operationally: you1ve shipped LLM-driven workflows with real acceptance criteria, not vibes 1 dynamic prompt/version/eval/orchestrate until the suite goes green.
• Are technically scrappy: comfortable scripting with Claude1s help in Python, querying SQL, calling GraphQL endpoints, and gluing tools/CLIs together (yes, even ffmpeg when needed).
• Think like QA: you design tests, hunt edge cases, and refuse to ship 3almost right.4
• Communicate crisply: tight specs, precise bug reports, clean dashboards, and docs others actually use.
• Are great at context switching and prioritizing among a large workload.
• Have an openness to change and a willingness to experiment with formats and platforms.
• Are not afraid to jump into any aspect of a project to fill a vacuum, no matter how big or small.