Deployment guides · Showcase platform

AI PM Bootcamp Capstone Gallery

A capstone gallery built on Cloudflare Workers so incoming AI PM Bootcamp students can see what prior cohorts shipped. One of three core capstone resources, alongside deployment guides and custom AI skills.

My role: Teacher Assistant and Capstone Lead. Built and launched the gallery as one of three core capstone resources (deployment guides, custom AI skills, and the gallery).
When: 2026 – present
Tools: Cloudflare Workers, AI Product Academy capstone program

The problem

Competing bootcamps had a showcase of student work. The AI PM Bootcamp did not, so incoming students had no way to see what a finished capstone looked like before starting their own, and the program had no home for the work its cohorts produced.

Approach

Identified the gap against competing bootcamps (they had a student-work showcase, the AI PM Bootcamp did not) and built the gallery to close it.
Built and launched the gallery on Cloudflare Workers: 18 finished prototypes in one place, behind student access.
Built and launched three core capstone resources in total: deployment guides, custom AI skills, and the gallery.
Primary point of contact for capstone teams building AI-native prototypes, owning the end-to-end capstone experience across cohorts.
Guided 190+ PMs through pressure-testing, building, and launching AI-native prototypes across three consecutive cohorts (191 students in total).
Hosted live Capstone AMA Office Hours and led Demo Day Pitch Sessions across cohorts, giving real-time technical guidance and evaluating final prototype presentations.
Used AI to mine recurring office-hours questions into new curriculum content.

Outcomes

Improved student retention across 191 students in three consecutive cohorts, raising per-cohort retention from 93.8% to 100% (95.3% average).
Gave every incoming cohort a concrete reference for what a shipped capstone looks like: 18 prototypes, where before there were none.
Closed a competitive gap against other programs with a resource that keeps paying off each cohort instead of a one-time fix.

What I learned

A shared showcase did more for motivation and polish than another round of feedback. Once students could see each other's finished prototypes in one place, the bar for what shipped meant rose on its own.

The reason I built this was simple. Competing bootcamps had a showcase of student work, and the AI PM Bootcamp did not. So I built a gallery for the program: one place where every cohort’s finished capstones live, so students coming in can see what previous cohorts built before starting their own. That is the main thing I wanted it to say: here is what people made, now go make yours.

The gallery is student-access-only, so it is not linked here. The work it shows belongs to the students who built it, not to my portfolio. What it shows on my side is the pattern: I spotted a competitive gap, scoped the smallest thing that closed it, shipped it on Cloudflare Workers, and it now serves every cohort that follows.

It is one of three core capstone resources I built and launched, alongside the deployment guides and custom AI skills. Across three consecutive cohorts and 191 students, I ran live Capstone AMA Office Hours and led Demo Day Pitch Sessions, which is where the pressure-testing happened. Students pitched, I pushed on the build and the positioning in real time, and the prototypes that came out the other side were the ones that earned a spot in the gallery. Per-cohort retention moved from 93.8% to 100% over those three cohorts (95.3% average), and recurring office-hours questions were mined back into the curriculum so each cohort started further along than the last.