ab-experimentation

Maintain a single living document for courses, goals, constraints, and accessibility so every tutoring and TA skill stays aligned.

Emit standardized progress pings during long-running work so the coordinator and the user know whether you are alive, stuck, or done — never go dark.

Read, react to, and operate GitHub Actions workflow runs and status checks like a careful operator — never blindly rerun, never blindly approve.

Before declaring work complete, run a structured self-assessment that catches the most common 'shipped half a thing' failures (missing tests, missing docs…

Design, implement, and maintain automated unit and integration tests that validate system behavior, prevent regressions, and ensure long-term reliability of…

Hand a long-running task from one agent to another (or back to a human) with a single, complete handoff packet so the receiver never has to reverse-engineer…

Canonical scaffold for new Syncolab skills.

Help learners and TAs navigate collaboration rules, citation, and AI policies without overriding the instructor.

Generate and iterate ad creative — headlines, primary text, descriptions, scripts, and image briefs.

Design rigorous A/B tests with clear hypotheses, sample size, and readout plans.

Optimize content to be cited and recommended by AI search engines (ChatGPT, Perplexity, Google AI Overviews, Claude).

Plan, instrument, and audit marketing analytics events (GA4, Mixpanel, Amplitude, server-side).