About

Andrea Cosentino

Senior Principal Software Engineer at IBM

I’m an open source enthusiast and Senior Principal Software Engineer at IBM, working on integration technologies. I co-lead the Apache Camel project together with Claus Ibsen, focusing on enterprise integration.

Most of what I do sits in the open: components, releases, security fixes, the boring maintenance work that keeps a project the size of Camel usable. This blog is where I write the parts that don’t fit in a commit message.

Open Source Leadership

🐪
Apache Camel PMC Chair Leading the Apache Camel project
🪶
Apache Member Member of the Apache Software Foundation
📦
Apache Karaf Committer Contributing to the OSGi runtime
🔧
Apache ServiceMix PMC Project Management Committee member

How I got here

I didn’t plan any of this as a career path. I picked up a component that needed fixing, then another one, and the rest followed. Looking back at what I was writing about, the arc is reasonably clear.

Current Focus

I’m currently working on expanding the Camel ecosystem through subprojects like:

Outside of Camel itself I contribute to Hawtio and to a number of smaller projects around the same ecosystem, usually because something broke and I happened to be the person looking at it.

A lot of the work is not feature work at all. Keeping a project the size of Camel healthy means release management, reviewing other people’s patches, triaging security reports, chasing dependency updates that nobody enjoys, and making sure the same route behaves the same way on Spring Boot, on Quarkus and on plain Java. None of that shows up in a changelog in a way anyone notices, but its the part that decides whether the project is still usable in five years.

The rest of my time goes on the boundary between the project and the people using it: answering questions on the mailing lists, writing up what changed and why, and trying to make the upgrade path from one major version to the next less painful then it needs to be.

Patents

Alongside the open source work I have spent a lot of time on patents: 41 inventions so far, 22 of them granted. Most were filed during the Red Hat years and a few more since moving to IBM. Almost none of it is solo work, it is mostly with Paolo Antinori and Leigh Griffin, and the full record is on my Google Scholar profile.

They fall into a few groups. A lot of messaging and event streaming: serializers, schemas, the claim check pattern, message sizing on mesh networks, compression for idempotent stores. Then secrets and key management, which is the same problem I ended up solving in the open with the Camel Vault property function. Then a long run of connected vehicle work: over the air updates, keyless entry, vehicle control on serverless functions. Lately it has been supply chain and AI: SBOM compliance, working out the minimal dependency update that clears a CVE, data provenance for foundational model training, and mediating access to quantum services.

The part that takes the time is not the writing. It is prior art search: reading through everything already published in an area before you can argue that the thing in front of you is actually new. I have done a lot of that, on my own submissions and reviewing other people’s, and it turns out to be a habit worth having well outside the patent process. It is the same instinct that tells you a brand new CVE is really the same bug shape you already saw two years ago.

Granted (22)

Filed and pending (19)
  • Auditable data provenance for training dataset prediction in large foundational models US 2026/0080305 · published 2026 with Leigh Griffin
  • Deployment of configuration files generated from serverless functions US 2026/0079720 · published 2026 with Leigh Griffin
  • Identifying minimal dependency updates to mitigate known CVEs US 2026/0079690 · published 2026 with Leigh Griffin and Paolo Antinori
  • Dynamic mediation of access to quantum services US 2026/0065123 · published 2026 with Leigh Griffin and Paolo Antinori
  • Mitigating vulnerabilities in a software program based on its usage US 2026/0037636 · published 2026 with Luca Burgazzoli
  • Event sourcing for quantum debugging and failover mechanisms US 2025/0377957 · published 2025 with Leigh Griffin and Paolo Antinori
  • Migrating ransomware activity of an operating system US 2025/0328650 · published 2025 with Paolo Antinori
  • Sizing service for cloud migration to physical machine US 2025/0208880 · published 2025 with Leigh Griffin and Paolo Antinori
  • Claim check mechanism for a message payload US 2025/0097181 · published 2025 with Paolo Antinori
  • Matching commands to attack patterns US 2025/0045395 · published 2025 with Paolo Antinori
  • Migrating secrets from a cloud environment to a local system US 2024/0297782 · published 2024 with Leigh Griffin and Paolo Antinori
  • Key rotation based on traffic state US 2024/0259197 · published 2024 with Paolo Antinori
  • User-customized vehicle control using serverless functions US 2024/0070258 · published 2024 with Paolo Antinori
  • Authenticating electronic key devices US 2023/0256936 · published 2023 with Paolo Antinori
  • Simplifying creation and publishing of schemas US 2022/0334811 · published 2022 with Paolo Antinori
  • Techniques for validating safety agreement compliance using SBOMs EP 4636583 · published 2025 with Leigh Griffin and Paolo Antinori
  • Mitigating ransomware activity of a host system using a kernel monitor EP 4567648 · published 2025 with Paolo Antinori
  • Intra-vehicle over-the-air updates based on install timestamp EP 4471578 · published 2024 with Paolo Antinori
  • Intra-vehicle over-the-air updates incorporating update compatibility EP 4471579 · published 2024 with Paolo Antinori

Interests

Beyond integration, I’m passionate about:

This Blog

This is my personal space where I write about open source, Apache Camel, programming, and technology in general. Lately it leans heavily towards security: what actually broke, and why, rather then the score somebody assigned to it.

Elsewhere

Best place to reach me is GitHub or LinkedIn. I read everything, I don’t always reply quickly.