Inkly.
Booking appointments without getting lost in direct messages.
An app, currently in development, to book and manage appointments without going through Instagram direct messages, where clients always end up getting lost. With a scoring system and a built-in chat.
- Mobile app
- Booking
- Scoring
- Chat
The client study.
Tattoo artists organise their appointments straight inside Instagram's private messages. As conversations build up, requests get mixed together and some end up forgotten, which leads to more missed appointments. Talking with several of them, the same need kept coming back: managing those appointments in a proper tool, outside the inbox.
An application built outside the inbox.
I designed an application to book and manage appointments, with a structured form and built-in messaging. The goal was to bring together, in one place, everything that concerns the artist and everything that concerns the client: the request goes through a clear form, the conversation continues in a dedicated chat, and every appointment is tracked without ever going back into the DMs.
A clear frame, then iterations.
I built the whole thing with the help of Claude Code, moving forward in iterations. Rather than writing everything in one go, I first set a clear frame and then let the AI work inside it: a documented project context, specialised agents for each step, and tests that validate the logic as it goes. Each iteration stayed small and verifiable before moving on to the next.
A CLAUDE.md file that defines the project's context and conventions.
Specialised sub-agents for research, planning and development.
Tests written upfront, on the business logic.
Frequent commits, one per task.
Scoring, refined by AI.
Each request receives a score that takes into account how complete the form is, how serious the message reads and the client's history. The AI adds a qualitative reading: it assesses the answers and the project description against what the artist considers a good client. The score then settles according to their own criteria.
The AI assesses how consistent the answers are with the artist's own definition of a good client.