Leonardo Pegollo

A district doctor in Milan. Community healthcare, complex patients and home care. In today's clinical work, data, prevention, proactive care and assisted code meet in the same field: not as a separate identity, but as ways to reduce friction and hold information, people and decisions together.

What I do for work

I work as a district doctor. I have always worked on community healthcare,complex patients and home care: the part of the system where care does not end with a prescription, but begins there. It is a profession made of pathways, networks, transitions between hospital and community, and time spent holding together pieces that elsewhere do not speak much to one another.

I recently completed a postgraduate programme in health and social-care management. It left me with a particular way of looking at processes: finding where a service breaks, understanding who really pays that cost, and trying to redesign it before the cost becomes chronic.

The thing I enjoy most

I am interested in the point where clinical operations, data and small applications can really speak to one another. Thinking about what to do, how to do it, and how to make it practical. At first for me. Then, if it works well, for anyone who wants a tool to manage their clinical work.

It is also the principle behind good service redesign: build small prototypes, try them with the people who work with them every day, and correct them without falling too much in love with the first idea. I do it with myself too, and I can describe it from experience.

Why a website

Because stopping at an application built only "for yourself" would be a shame. When you study, you find useful, sometimes wonderful reflections on how medical applications are conceived, developed, structured,distributed and sometimes misused. Privacy, app topology, data flows, rules and connections to regional infrastructures: the less visible part of digital tools in healthcare.

I would like to write about what it means to provide healthcare when the time between a question, data and a prototype becomes much shorter. It is good news only if it stays inside a method: humility about what we can do, review of the results, and care not to turn every tool into content to produce. Clinical data are not opinions. Topology choices are not matters of style. Rules are not obstacles to the product: they are the product.

What you will find here

For now, the MediFlow presentation. From here on, the blog will collect working notes on privacy, clinical-app topology, flow governance, participatory prototyping, data quality, SISS/FSE integration, and a few freer pieces on how community healthcare is studied, decided and sometimes got wrong.

No fixed schedule. No clickbait. When I have something to say that stands up to rereading, I write it.

Bounded collaborations

I am interested in bounded collaborations and European roles around AI in health and the public sector, especially deployment and evaluation: concrete problems, clear perimeters, traceable sources and explicit checks. I am not claiming a general mandate, or presenting MediFlow as adopted or validated; I am interested in contributing where the work can be defined, tested within its scope and handed back with its limits.

Tools I use

SQL, Python and R to read data and put things in order. TypeScript, Astro and other web tools when they help. I write code myself and also use modern assistants, but with one simple rule: the result must stand up to review and serve a purpose. If it only produces volume, I am not interested.

How to write to me

Three channels, in decreasing order of formality: email, X, GitHub.

Get to the point: two lines work as well as twenty, as long as I understand what you are looking for. I reply within a few days when I am not caught up in day-to-day clinical work.

I write this page in the first person and then dry it out a little. If you find an inaccuracy in the professional details, write to me.