The work
What I built and how it works
PitchMate is three products in one sequential flow. Video-based knowledge modules with embedded quizzes. An AI roleplay simulator. A structured performance analysis layer. Each one feeds the next.
Training flow — 8-stage curriculum, knowledge stages gated before pitch practice
Every video module pauses at key moments to surface a knowledge check. A mandatory correct answer before the video continues. The quiz gate isn't just at the end — it runs through the entire learning phase, preventing passive watching.
Interactive video — auto-pause at key moments, mandatory correct answer to proceed
Once the knowledge check is cleared, Rhea appears. Rhea is the AI prospect — she plays a student, a budget-conscious parent, or a working professional depending on the scenario. She raises real objections, adapts dynamically to the rep's responses, and doesn't accept surface-level answers. Auto-repetition runs until the rep achieves mastery score.
AI Roleplay Simulator — Rhea plays a live prospect, adapts to responses, throws real objections
After every session, the rep gets a structured performance breakdown — pitch evaluation across five dimensions, specific objection handling analysis, manager comments, and a recommendation for the next session. The analysis is where improvement actually happens.
Post-session analysis — pitch score, objection handling breakdown, manager comments, next session suggestion