Product / UX
Course recommendation funnel
60–90s Flask experience recommending up to 3 courses by interest, time and modality, with store CTA.
Context
Enrollment-style recommendation flow: turns indecision into a short actionable menu without expensive AI—deterministic rules and Flask session.
The challenge
- 1Complete the funnel under 90s.
- 2Explainable recommendations.
- 3CTA to a real store without rebuilding checkout.
The approach
01
Short wizard
Welcome → interest → time → modality → menu.
02
Rules
Deterministic, cheap, auditable engine.
03
Store handoff
CTA into external course catalog.
Architecture
5 guided steps → rules engine → ≤3 course menu → external store CTA.
System layers
- 01Multi-step templates
- 02Flask session
- 03Rules engine
- 04Course model
- 05Store CTA
Applied stack
- Python
- Flask
- Jinja
- Session
- UX funnel
Design decisions
Rules vs LLM
Explainability and zero cost on a short flow.
Short menu
Choice overload kills conversion.
External checkout
Do not reinvent payments/enrollment.
Outcomes
- 5
- Funnel screens
- ≤3
- Recommended courses
- 90s
- Target duration
Key points
- Conversion through clarity, not a chatbot.
- Near-zero marginal cost.
- Reusable pattern for other catalogs.
Fit / how to hire
Hire this if your growth/product team needs measurable funnels that convert through clarity, not an expensive LLM, with a clean handoff to your store. Discovery → prototype → instrumentation.
Book a technical call
