Gianluca Zompa, software developer

Private by architecture.

I design and engineer apps for iPhone, iPad, Mac and Apple Vision Pro. Your data stays on your device because of how each app is built, not because of a promise in a privacy policy.

Independent and full-time, writing Swift since iOS 8. When a new framework comes out and an app needs it, I study it and build something real with it: Bookslot on SwiftData and Magic Diary on Foundation Models, both before any guide covered real use cases.

Gianluca Zompa on a sunny terrace, with the hills of Vairano Patenora behind him
In Vairano Patenora, southern Italy, where it’s now.

What each app knows, and where it keeps it.

Four apps, four different problems, one rule: your data is yours. For each one, here is what it stores, what leaves your device and how the code makes sure of it.

Magic Diary: The journal: entries by day, with the feelings tagged Magic Diary: The main view, with the sea at the bottom Magic Diary: An entry retold as a tale

iPhone, iPad, Mac and Apple Vision Pro

Magic Diary

Turn your days into tales.

A safe place for your days: support when they’re hard, a place to keep them when they’re good. Everything you write, and everything the AI does with it, stays on your device.

Get it on the App Store How Magic Diary works

Where your data goes

Stays on your device
Entries, moods, photos, voice memos and stories.
Written by
Apple’s on-device language model. Your words never go to an AI server, mine or anyone else’s.
Synced through
Only your own iCloud account, with Apple’s CloudKit.
Pomodrop: A session running: the sun on the dial, the water risen Pomodrop: The next session, with the duration it suggests and why Pomodrop: Guided breathing

iPhone, iPad, Mac and Apple Vision Pro

Pomodrop

A Pomodoro timer that adapts to you.

A focus timer that learns from your sessions instead of imposing a rhythm, and helps you rest between them.

Get it on the App Store How Pomodrop works

Where your data goes

Stays on your device
Sessions, tasks, diary entries and your focus history.
Suggestions come from
Your own sessions, calculated on your device. No AI, no server.
Account
None to create.
Bookslot: Your library by list and status Bookslot: The reading timer Bookslot: Saved quotes

iPhone

Bookslot

Your library, and the time you spend reading.

Your book collection and your reading, in one place, so you always know what you own, what you’re reading and how much you read.

Get it on the App Store How Bookslot works

Where your data goes

Stays on your device
Your library, reading sessions, quotes and purchases.
Leaves your device
Only the title, author or ISBN you search, sent to Google Books, or to Open Library for barcodes, to find the book.
Account
None to create.
Key Press Counter: The menu bar panel with today’s count and sky

Mac, in the menu bar

Key Press Counter

How much you typed. Never what.

How much you typed, never what. From that number alone, it shows you the shape of your working day.

Get it on the Mac App Store How Key Press Counter works

Where your data goes

What it reads
A running total of key presses that macOS already keeps. The app never receives a single keystroke.
Password fields
Keys typed while a password field is active are left out.
Permissions and network
None. No Input Monitoring, no internet connection.

How I work

  • I study before I build.

    When a new framework comes out and an app needs it, I learn it properly and then build something real with it. Bookslot runs on SwiftData and Magic Diary on Foundation Models, both started when there were no guides for real use cases, only documentation and samples.

  • Privacy is part of the architecture.

    If an app doesn’t need your data on a server, it never sends it. Where something has to leave your device, I tell you exactly what and why, as you can read for each app above.

  • A product, not a funnel.

    You pay once, or start free and unlock everything with a single purchase. No subscriptions forced on you, no ads, no dark patterns.

  • Every app gets its own interface.

    A diary that turns pages into tales, a timer that follows the sun, a library for collectors, a sky over your keyboard. Each one is designed around what it does, never assembled from generic parts.

Now November 2026

Building
Foliolect. A native Mac app for running language models locally, and what I work on full-time right now. Get an email when it’s out.

Self-taught means I chose my teachers.

I started with Marcello Catelli’s Swift Academy course. Then, before launching any product of my own, I chose to study properly: the first book was Matt Neuburg’s Swift per iOS 8, and one book led to the next.

Below is part of the paper shelf: algorithms and data structures, operating systems and networks, languages and paradigms, compilers, machine learning and deep learning. About twenty more books are digital.

  • The Art of Computer Programming, Vol. 1
  • The Art of Computer Programming, Vol. 2
  • The Art of Computer Programming, Vol. 3
  • The Art of Computer Programming, Vol. 4A
  • Introduzione agli algoritmi e strutture dati
  • Data Structures & Algorithms in Swift
  • Code
  • Architettura dei calcolatori
  • Sistemi operativi
  • Compilatori: principi, tecniche e strumenti
  • Linguaggi di programmazione
  • Swift per iOS 8
  • Programming iOS 8
  • Modern Concurrency in Swift
  • Design Patterns
  • Clean Architecture
  • Il Pragmatic Programmer
  • AI and Machine Learning for Coders
  • Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow

The whole shelf, by subject

Writing

The engineering behind each app, decision by decision. The first piece explains how Magic Diary retells a diary entry with Apple’s on-device model: the structured output, the prompts, and what failed along the way.

It’s being written now. Subscribe and it reaches you the day it’s published.