01
Overview
The marketing and onboarding site for an AI-driven trading platform, including a surface that recommends courses to each visitor.
02
Problem
Turn first-time visitors into onboarded users, and put the right learning path in front of each person rather than a generic catalogue.
03
My role
- Architected the marketing and onboarding site.
- Built the personalised course-recommendation interface.
- Implemented the front end in vanilla JavaScript and TailwindCSS on a PHP backend.
04
Architecture
- FrontendHand-written HTML, CSS and JavaScript with TailwindCSS — no framework.
- BackendPHP.
- FeaturePersonalised course-recommendation surface in the onboarding flow.
05
Technology
Interface
- JavaScript
- HTML/CSS
- TailwindCSS
Backend
- PHP
06
Engineering challenges
C1
Framework-free interactivity at scale
A full marketing and onboarding experience with no framework to lean on for structure.
C2
Personalised recommendations in a lean stack
Per-visitor course recommendations without a heavy client runtime.
07
Solution
- S1
Framework-free interactivity at scale
Structured vanilla-JS modules for the onboarding flow so behaviour stayed maintainable as the site grew.
- S2
Personalised recommendations in a lean stack
Drove the recommendation UI from the PHP backend and kept the front end progressive and fast.
08
Results
- User engagement — first quarter
- Course sales — first quarter
09