Eight services in three areas, one goal: software and infrastructure your team can evolve on its own.
MOD
Artificial intelligence that works with your data and tools, with limits and controls.
I design AI agents that work with a company's tools and data: document search (RAG), tool use (function calling, MCP), flows with human approval. Every agent gets minimal permissions, capped costs and a measurable evaluation before it goes to production.
EN
Software and systems that are easier to change: code, migrations, connected systems.
I make existing software easier to understand and change. Tests first around current behaviour, then small steps: clear module boundaries, fewer dependencies, less technical debt. The system stays in production the whole time.
For each application I pick the right strategy among the "7 Rs" of migration (rehost, replatform, refactor, repurchase, relocate, retain, retire). Replatforming moves the application with limited optimisations: containerisation, managed databases, an updated operating system. Refactoring rethinks the architecture and costs more: I only propose it when it is needed.
An independent review of architecture, cloud costs, security and reliability. It ends with a short document: what is fine, what is at risk, what to do first. It is the simplest way to start working together.
I design and build home automation and IoT systems: sensors, actuators, protocols (MQTT, Zigbee, Modbus), local automations, data collection and dashboards, cloud integration. Where it makes sense I bring the same cloud native discipline to the edge: safe updates, observability, lightweight containers and compact clusters. Security, privacy and working without internet come first.
TRA
Platforms and automation that take everything to production repeatably.
I design and build Kubernetes clusters and cloud native platforms: architecture, networking, storage, observability, security, upgrades. It fits teams starting from scratch and teams with a cluster nobody dares to touch.
I move applications to Red Hat OpenShift Container Platform and make them run well: cluster architecture, installation and upgrades, security (SCC, RBAC, network policies), operators, pipelines and GitOps, observability. Also useful for teams coming from plain Kubernetes who must adapt to OpenShift, or from virtual machines who want to containerise.
I turn manual steps into repeatable processes: release pipelines (Jenkins and others), configuration with Ansible, infrastructure as code, GitOps, identical environments from development to production. The goal is to make releasing boring.
SMEs, startups and engineering teams with software or infrastructure to modernise, or that want to put AI agents in production without first building a dedicated department.
No. I am based in Pavia and work on site in Lombardy by appointment; the rest of the work is remote, across Italy and Europe.
Replatforming moves an application to the cloud with limited optimisations (for example containers or a managed database) without changing its architecture. Refactoring redesigns it to take advantage of the cloud: more benefit, higher cost, done only when needed.
It depends on how it is built. I work with least-privilege permissions, data separation, logs of what the agent does and human approval for sensitive actions. There is always a measurable evaluation before production.
It depends on the scope. After the first call you receive a written proposal with timing and cost. A short assessment is the simplest way to start.
Yes: code, configuration and documentation live in your repositories and your accounts. The goal is that your team can maintain them alone.