© Mapbox, © OpenStreetMap

Speaker

Roberto Franchini

Roberto Franchini

Principal Software Engineer @ ArcadeData

Principal Software Engineer @ ArcadeData

Turin, Italy

Actions

I'm a passionate, seasoned, wise, software craftsman with more than 25 years of experience.
I had the opportunity to work on a wide range of application types, from the monolith to the microservices, from fast-data architecture to database engine development, from manual deployment on bare-metal servers to cloud-native.
I like to share my knowledge as a regular speaker at tech conferences: JavaZone, VoxxedDays, Codemotion, WorkingSoftware, CloudConf, NoSqlDay, AgileDay, ItalianDevopsMeeting (IDI), BetterSoftware.

I'm a passionate, seasoned, wise, software craftsman with more than 25 years of experience.
I had the opportunity to work on a wide range of application types, from the monolith to the microservices, from fast-data architecture to database engine development, from manual deployment on bare-metal servers to cloud-native.
I like to share my knowledge as a regular speaker at tech conferences: JavaZone, VoxxedDays, Codemotion, WorkingSoftware, CloudConf, NoSqlDay, AgileDay, ItalianDevopsMeeting (IDI), BetterSoftware.

Area of Expertise

  • Media & Information

Topics

  • Software Architecture
  • Graph databases
  • LLMs & Agentic AI

To cloud, or not to cloud?

Today, the cloud has become the preferred choice for applications requiring elasticity, scalability, and cost efficiency. But what if these benefits could be achieved without the complexities of cloud infrastructure?
Kamal aims to democratize cloud-like deployment experiences.
Web applications can be easily deployed across various platforms, from VPS to cloud providers to self-managed bare-metal servers.

The Bug Arms Race: Surviving AI-Generated Bug Reports in Open Source

What happens to an open-source project when both sides get AI?

At ArcadeDB, a small team adopted an AI-native workflow, and our output jumped 28×. Visibility followed, and so did a 13× jump in AI-generated bug reports. Finding bugs had become free. Verifying them had not.

This is a data-driven account of that arms race, taken from the public ArcadeDB repository. You'll see what broke: triage bandwidth, maintainer focus, and the signal-to-noise ratio in the issue tracker. You'll also see what we built to survive: intake policies for AI-generated reports, a test net strong enough to confirm or reject a claim automatically, and an agent pipeline that uses AI to fight AI.

For maintainers, tech leads and engineers adopting AI-assisted development who want to scale output without drowning in it.

Break It Before Production Does: Chaos-Testing a Raft Database Cluster

Building a distributed database is hard. Proving it survives real-world failures is harder. In this talk we open the ArcadeDB test lab and show how we break our own Raft-based high-availability cluster on purpose, on every change and every week.

You'll see how we use Testcontainers and Toxiproxy to spin up real multi-node clusters and inject the failures production loves: leader crashes, rolling restarts, network partitions, split brain, latency and packet loss. Then we go further: chaos tests that run unattended for one to several hours, hammering the cluster with load while faults keep coming, and verifying that every replica converges to consistent data.

Distributed bugs are notoriously hard to explain, so we built animated diagrams of every scenario. You'll watch leader elections, log replication and partition recovery step by step, exactly as the tests run them.

For Java developers and architects who build or run distributed systems and want to sleep better at night.

Integration tests: Testcontainers to the rescue

Integration testing is hard, and often teams are tempted to do it in production. Testcontainers allows writing meaningful integration tests spawning Docker containers for databases, queue systems, kv-store, other services. Don't mock your database with fake data anymore, work with real data!

Kotlin social graph

Ingredienti. Lo stream di Twitter, un graph database, reactive streams. Amalgamate il tutto con abbondante uso di kotlin. La ricetta verrà preparata sul momento in perfetto stile street food.

Working Software 2026 Sessionize Event

June 2026 Reggio nell'Emilia, Italy

Incontro DevOps Italia (IDI) 2025 Sessionize Event

March 2025 Bologna, Italy

Working Software 2023 Sessionize Event

June 2023 Milan, Italy

Working Software Sessionize Event

July 2019 Milan, Italy

DevOpsHeroes 2018 Sessionize Event

October 2018 Parma, Italy

Milan Kotlin Community Conf Sessionize Event

June 2018 Milan, Italy

Roberto Franchini

Principal Software Engineer @ ArcadeData

Turin, Italy

Actions

Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.

Jump to top