Luby is Certified in the Claude Partner Network. Our engineers hold Claude certifications, and we are building from here.
What the Claude Partner Network is
The Claude Partner Network is how Anthropic works with the firms that put Claude into production for customers. The Services Track recognizes partners for building a certified delivery team, delivering Claude use cases for clients, and proving the work through public customer references. Partners advance through Certified, Select, Preferred, and Global Premier tiers.
You can learn more about the program at claude.com/partners.
The team behind the badge
This recognition belongs to our engineers. Every Claude certification behind it was earned by someone on our team who chose to keep learning: studying together, testing what they learned on real engineering problems, and helping the next colleague get there.
That is how we approach new technology at Luby. We do not wait for a tool to become mainstream before learning it in depth. We build the knowledge inside the team first, so that when a client needs it, the people doing the work already know where it helps, where it falls short, and how to run it safely in production.
Certification is a starting point, not a finish line. Our people hold Claude certifications, and we are building from here.
How we put Claude to work
We use Claude, powered by Anthropic, where it changes how software gets built and run. Two areas carry most of that work today.
AI agents in production
We build autonomous agents integrated into our clients’ operations, on a modular and observable architecture, using Claude Code and MCP. An agent is not a prompt wrapper. Each one is born owning an outcome: a business metric and a cancellation point are defined before the first line of code. In production, it runs with governance, immutable logging and a kill switch. See how we build AI agents.
AI legacy modernization
The hardest part of a legacy system is rarely the old technology. It is the business rules that live only in the code, written by people who have since moved on. We use AI to read that code and recover the documentation that does not exist, then migrate in phases, with the new system running alongside the old one. The operation never stops. See our approach to legacy modernization.
In both cases, agents take on the repetitive work: reading code, drafting specifications, writing tests, documenting. Senior engineers define the work, review it and approve every step.
What it means for our clients
For the companies we work with, this comes down to three things.
- A certified team doing the work. The engineers who design and build your Claude use cases are the same people who studied and certified on it.
- Engineering before the model. Most AI projects that disappoint fail on integration, data and architecture, not on the model. We start there: what the system needs to read, what it is allowed to change, and how the result is measured.
- A human in control. Agents execute; engineers decide. Every stage keeps human checkpoints, traceability of what the agents did, and governance over security and data.
We are a product engineering firm with offices in Miami and a delivery center in Brazil, with more than two decades building financial, regulated and high-stakes systems. ISO 27001 and ISO 27701 certified.
Building from here
To everyone on the team who made this happen: thank you.
If you are evaluating where Claude fits in your operation, from agents running real workflows to modernizing the systems underneath them, we would be glad to talk it through.