Skip to content
PERRY DOLMANS AI BUILDER & CREATIVE TECHNOLOGIST
WORKLABNOTESABOUT Get in touch

← BACK TO WORK

MOBILE APP · PHOTOGRAPHY

LAB76

A slow photography experience for Android.

STATUS
ANDROID BETA
ROLE
Concept, product, design & development
TOOLS
Claude Code · Android · AI image tools
YEAR
2026

A slower digital camera experience inspired by analogue film and delayed development.

01

The idea

Phone cameras are technically excellent and emotionally flat. We shoot hundreds of frames we never look at again. LAB76 asks a different question: what if taking a photo cost something again, attention, patience, a little restraint?

02

The problem

Unlimited storage and instant preview killed the moment of anticipation. There is no waiting, no surprise, no roll to finish. I wanted to design that friction back in on purpose, and see whether constraint could make everyday photography feel meaningful again.

03

The approach

I built LAB76 in phases with Claude Code: concept and naming first, then the fictional film-stock system, then the camera, then the develop-and-wait loop. Each phase shipped something testable before the next one began. Nothing moved forward on faith.

04

My role

Everything. Concept, product decisions, brand and packaging, interface design, Android implementation and the AI-assisted asset pipeline. A solo build, which is exactly what the current generation of tools makes possible.

05

How AI was used

AI did not make the decisions. It removed the distance between an idea and a working version of it.

Naming and brand language
Designing fictional film-stock characteristics
Android implementation with Claude Code
Interface and icon design
Generating sample “developed” frames
Writing and structuring the documentation and the software
06

Key decisions

The hardest call was the wait. A real delay risks people uninstalling, so a roll develops over hours, not days, and a notification brings you back. Fixing the roll length made every frame feel deliberate instead of disposable.

/ THE SYSTEM

The core loop — everything else supports it

01
Load film stock
02
Shoot a roll
03
Seal & develop
04
Hours, not seconds
05
Open the roll
/ SCREENS & PROTOTYPES
07

What worked

The develop-and-wait loop. Testers reported checking back with genuine anticipation, the exact feeling the product was designed to create. The fictional film stocks gave the brand room to grow.

08

What did not work

The first version showed images instantly behind a fake “film grain” filter. It felt like a gimmick and I scrapped it. Skeuomorphic shutter sounds tested as annoying rather than charming, so they went too.

09

What I learned

Constraint is a feature. And a solo builder with AI can now carry a product from a single sentence to a private beta, brand, app and assets — without a team.

CURRENT STATUS

In private Android beta with a small group of testers. The film-stock system and develop loop are stable; I’m tuning timing and onboarding.

NEXT STEPS

A wider beta, more film stocks, and a shared “darkroom” where rolls can be revisited. A full build log is coming to Field Notes.

Interested in how this was built?

Get in touch See more work