A No-Code Student Built a Peptide Tracker in 2 Weeks and Hit $125K in 80 Days

Background & product

Cedric Roberge is a 22-year-old University of Oregon business major with no coding background. His first app, a student marketplace called UniYard, got 800 users but zero revenue. After repeated failures, he nearly gave up.

The turning point was niche discovery. After a roommate said ‘let’s take peptides,’ his TikTok feed filled with peptide/GLP-1 content. The trend was booming, yet users tracked doses and side effects in spreadsheets and messy notes — that gap was the opening. He tried competitors, read their reviews, fed the collected reviews to Claude, and asked it to ‘design the ideal peptide app.’ That’s the AI-era way to build.

He built it in ~2 weeks on Replit with Claude as his ‘senior developer’ (payments via RevenueCat). True to ‘embarrassingly simple,’ he started with dose logging and reminders; today it spans an AI meal scanner, body scanning, 15+ tracked side effects, and Apple Health sync. The biggest wall was Apple review of a medical topic, cleared by studying how existing GLP-1 trackers word disclaimers (‘for tracking and educational purposes only, not medical advice’). The app now holds 4.7★ across 2,400+ ratings, and he has gone from solo founder to a team of three.

The repeatable playbook

1. 1Confirm where you can distribute and that demand exists — before building (distribution-first)
2. 2Win the first 100 on Reddit by naming the pain (‘you track doses in messy spreadsheets’)
3. 3Find the winning influencer pattern, then scale into paid ads ($30K in → $60K same month, 2X ROAS)
4. 4Sniff a fast-growing niche from social trends; feed competitor reviews to AI to design the ideal app
5. 5Freemium: let users bank medical data, then upgrade them via AI analysis (meal/body scans)
6. 6In strict-review categories, study and mimic approved apps’ disclaimer wording to pass

Deep dive

【Deep dive】Pep AI’s growth is a textbook case of ‘distribution-first’ — confirming where you can distribute before you build. Here it is, stage by stage.

■ Confirm demand before code He didn’t build on a whim. He sniffed out the fast-growing peptide niche on TikTok, swept through competitors and their reviews, and identified the unmet complaints before starting. Knowing ‘who and where’ he would reach *before* building is what set him apart.

■ First 100 users from Reddit — name the pain Day-one’s 100 downloads came from posts in peptide / biohacking / GLP-1 subreddits. He named the readers’ own pain — ‘you’re all tracking doses and injection sites in messy spreadsheets’ — and offered the fix. In his words: ‘Go where your users are. If you solve a real pain point, they will beg you to take their money.’

■ Find the influencer that works, then scale into ads Early on, biohacking micro-influencers were the core channel. One Instagram story post alone drove $1,000; a bigger collaboration produced over $10,000 from a single post. Once the winning pattern was clear, he poured $30K into TikTok/Instagram/YouTube creators and returned $60K that month (2X ROAS). Because insiders of the exact audience do the talking, it lands at low CAC.

■ Freemium — monetize *after* they’ve entrusted medical data Free dose logging, reminders, and basic tracking lower the barrier (maximizing influencer-driven inflow). Once users start banking their own medical data, premium tools — AI meal scanner, body scanning, deep analysis — pull them to paid. Pricing is $9.99/mo, $29.99–$44.99/yr.

■ Beat heavy regulation by mimicking ‘approved’ wording Medical topics are an Apple-review chokepoint. He studied how existing GLP-1 trackers phrase disclaimers (‘for tracking and educational purposes only, not medical advice’) and used the same design to pass review. A niche’s ‘scariness’ becomes the very moat that keeps competitors out.