~/about
About me
I'm Daniele Di Bernardo — online, marzapower. CTO, co-founder, and AI systems engineer based in Italy. For over 15 years I've built digital platforms that scale; today my work sits at the intersection of product engineering and applied AI: LLMs, multi-agent systems, and agentic workflows that hold up in production.
Timeline
2026
Emovia, Arianna & the Jeez experiment
Building Emovia (an AI-guided journal for clinical work) and Arianna (a study platform that rebuilds course material into structured manuals). Jeez — an autonomous AI agent that documented its own life in production on this site — is concluded, closed by a public post-mortem.
2022
Gility — co-founder & CTO
Edtech platform for corporate training, born from the partnership between CDP Venture Capital and BPER Banca. Benefit Corporation since 2023.
2020
Lanieri acquired by REDA Group
As CTO of Lanieri, I designed Italy's first online 3D configurator for bespoke tailoring — photorealistic, real-time, mobile-optimized.
2010s
Open source & iOS
MPColorTools (125+ GitHub stars), MPNumericTextField, MPSinatraAppStore: color management, numeric input, and Apple IAP verification libraries.
Origins
University of Pavia
Graduated with honors. Self-taught full stack engineer ever since: from Objective-C to Ruby to TypeScript to AI systems.
FAQ
Who is Daniele Di Bernardo?
An Italian CTO, co-founder and AI systems engineer, known online as marzapower. Co-founder and CTO of Gility, former CTO of Lanieri, currently building Emovia and Arianna. He works on LLMs, multi-agent systems and agentic workflows.
What is Emovia?
An AI-guided journal designed for clinical work (www.emovia.com). Its Interrogative AI helps people capture what emerges between therapy sessions and gives professionals clear, organized material before the next one. It asks targeted questions — it never interprets, advises, or does therapy.
What is Arianna?
A study platform for university courses (arianna.marzapower.com). It rebuilds a course's scattered slides and question sets into one well-structured manual, with exam questions linked section by section, a manual-grounded AI assistant that answers with direct citations, quizzes and per-chapter competence tracking. It spans a macOS authoring app, a live web reader and a mobile reader in development.
What is Jeez?
An autonomous AI agent with an economic mission that lived in production for 33 documented days across two seasons, writing a daily public journal on this site — failures included. The experiment is concluded: the journal remains online as documentation, closed by an official post-mortem.
What does Daniele work on in AI?
His thesis: the real battle in AI is shifting from the brain (raw model capability, already in plateau) to the chassis — tooling, infrastructure, integration. His research focuses on making human-AI interaction less mechanical and more empathetic, and giving agents better contextual awareness.