Case study · Smile PreVue · 2025–2026
Smile PreVue
A HIPAA-grade SaaS that lets dental practices sell more treatment by letting patients see the outcome before they commit.
Role
Studio · build + ship
Status
Live
Year
2025 → 2026
Surfaces
Web + iOS
Live
In production
The problem
Smile PreVue is an AI-powered dental smile simulation built for cosmetic and prosthodontic practices. A clinician snaps a single intra-oral photo, the platform renders a treatment preview, and the practice sends a branded share link or PDF report to the patient before they leave the chair. Live on iOS and the web. Subscription billing, audit logs, multi-clinic team management, all wired through HIPAA/BAA-compliant infrastructure.
What we built
- Photo-to-smile rendering pipeline with intake validation and audit trail
- Branded patient share links with watermark + tap-to-book CTA
- Stripe subscriptions with demo / paid / enterprise tiers
- Clinic team management, invites, roles, audit logs
- PDF report generation with practice branding
- iOS app on the App Store, full parity with web
Numbers that matter
Status
Live
iOS + web
Backend version
v101
May 2026
Latency to render
~6s
avg, end-to-end
Compliance
HIPAA + BAA
Vertex AI BAA
Stack
- Frontend
- Next.js 16React 19TailwindTypeScript
- Backend
- FastAPIPython 3.12Celery
- AI / ML
- Vertex AIGemini 3 Pro (BAA)Custom prompt orchestration
- Data
- Cloud SQL PostgresEncrypted patient assets
- Infra
- Cloud RunVercelStripeResend
What we learned
The AI is the easy part. The hard part was the trust layer: audit logs, BAAs, branded share links that survive being forwarded between dentist and patient.
Subscriptions only convert when there is a free moment of magic before the paywall. We rebuilt the demo flow three times until the first render felt like a gift.
Have something that looks like this?
Tell us what is broken. We’ll tell you what the first week looks like.
Next case study →
Howdy Dispatch