I'm Luca Di Domenico: fractional CTO, full stack developer, and instructor. Founders, startups, and companies hire me to turn a product idea into a web or mobile platform that works, holds up under load, and keeps evolving over time.
I work solo, with a method built to handle complex projects. First I get clear on the product, the constraints, and the priorities. Then I translate all of it into working specs, build with AI agents, and document the technical choices that matter. The result is a project that stays readable, maintainable, and in your hands, never code that only I can make sense of.
I've been building software for over ten years. I started as a full stack developer with a focus on application security, and from there I've built web apps, mobile apps, dashboards, APIs, and SaaS platforms for clients, startups, and projects of my own. That background is why, when a new problem lands on my desk, I already know where to look and what's likely to break.
Agentic coding with Claude Code changed the way I work, and not only because of the speed: the real leap is in control. I write precise specs, think through the edge cases before I type a single line, check every output, and document the decisions. AI agents multiply what I produce, while the project stays driven by someone who knows exactly what they're doing.
This is the whole point: AI delivers solid results only inside a clear process. That's why I work spec-driven. I define what the software has to do and how it should behave, the agents build it, and I review every step. It's why one person, working this way, ships as much as a team, without the overhead of having to coordinate one.
I build web and mobile platforms, dashboards, APIs, and SaaS. My work goes well beyond writing code: I start from the problem, help define the product, and build a solution that can be used, tested, and improved over time.
Working solo lets me keep everything consistent across the whole process: product, architecture, development, technical UX, integrations, and code quality.
I share my method through YouTube, talks, technical content, and AI Builders Italia. I care about showing the real work: how I set up a project, how I write the specs, how I use AI agents, how I review code, and how I handle errors, limits, and ambiguity.
My goal is to help other builders, developers, and founders use AI in a way that's more concrete, less chaotic, and closer to the needs of a real project.
I support founders, startups, and companies as a fractional CTO, full stack developer, or trainer. I can help you build a product from scratch, review a technical direction, improve your development process, or train your team on the practical use of AI in software.
I built Podscape on my own: a mobile app for tourists, a web app for artists, an admin dashboard, and an API. It's a platform for geolocated musical and cultural tourism, designed to connect places, routes, and audio content tied to the local area.
I worked on Aurea, a crypto mobile app with an integrated wallet, flows for buying, selling, and swapping, and access to DeFi features. A complex mobile project, with technical, UX, and compliance constraints to manage carefully.
I built Filterly as a complete digital platform, taking care of user experience, application logic, architecture, and development. It's a good example of my approach: start from a specific problem and arrive at a solution that's clear, usable, and ready to evolve.
In 2026 I started bringing agentic coding to the stage as well. In June I spoke at CTRL SHIFT, at the Parthenope University of Naples, with a talk on how the developer's role changes as AI agents arrive.
I like talking about technology from a practical starting point: what changes in day-to-day work, which skills become more important, and how you build software when AI truly enters the development process.
On my YouTube channel I show how I use Claude Code, agentic coding, and technical specs on real projects: from planning to development, all the way to reviewing the code the AI generates.
A practical guide to getting started with Claude Code in 2026 and understanding how to fit it into a development process that's more orderly, more controllable, and more productive.
A practical breakdown of the most common mistakes people make when they use AI without specs, context, and technical control, with a focus on the method I use in my own projects.
The process I use to check, guide, and correct the code Claude Code generates, cutting down on ambiguity, bugs, and errors that are hard to catch right away.
I founded AI Builders Italia to bring together people who want to use AI seriously in software development. A community with 1,000+ builders, a Pro membership, a Discord, and hands-on courses on Claude Code and agentic coding.
