Most AI Agents Never Ship. Here's How We Did.
Join me in SF to see how we scaled ZTRON to production with limited resources
The truth about AI Agents? There are too few of them in production.
That’s why I will present at an in-person event in San Francisco on how we designed, shipped and scaled ZTRON, a vertical AI agent that I’ve been working on in the past year that is currently running in production.
We won’t talk about any imaginary use cases, PoCs, or scenarios where you have billions in investments and don’t care about costs.
Instead, we will present something real, which is 100% hands-on, based on how we managed to ship an AI agent to production (that actually has real users) with limited resources.
A scenario that most of us face when building a new product.
👋 Also, it will be a fantastic opportunity for me to meet in person people from the San Francisco AI community. The energy around AI in this city is fantastic. The best I’ve seen so far.
The Event
It will be a quick but packed event, hosted by Stephen Ribbon, CEO of ZTRON.
🤫 P.S. This is the stealth start-up I’ve been working on. We are finally going live!
There will be three stages:
Presentation by Qian Li, PhD, co-founder and architect at DBOS, Inc., on how to build AI Agents that actually work in production using DBOS, which allows you to easily leverage durable workflows, orchestration, queues and scheduling strategies.
Presentation by Jona Rodrigues and me showing how, at ZTRON, we managed to design and scale our vertical AI agent for financial services using DBOS by being smart and flexible, not throwing in Kubernetes at each problem.
Panel on building production AI
Why is this exciting?
The reality is that if you want to be flexible while still scaling and building a robust AI solution, you don’t need any fancy AI frameworks or infrastructure tools to hinder you.
That’s why we invited DBOS to present beside us, as it’s one of the few tools that let you easily scale your AI agents without adding any extra complexity. They serve as replacements for AI Frameworks such as LangGraph, orchestrators like Airflow, and background workers like Celery. Amazing, right?
In more detail, we will present:
A vertical AI agent solution that’s already working in production.
A flexible cloud architecture that doesn’t parallelize you, but lets you quickly iterate and adapt. Critical for AI start-ups.
Showing hands-on how you can build end-to-end agentic solutions with a simple, but powerful tech stack that lets you innovate: AWS, FastMCP, Postgres and DBOS.
What works and doesn’t work when pushing agents to production. Spoiler alert. Most techniques out there don’t work.
Join the Event
The event is in-person on November 5th, from 4:00 PM to 6:00 PM, at the Antler VC SF Office in San Francisco.
Everything is free of charge.
If you want to chat at a casual beverage about AI solutions in person with me, the ZTRON and DBOS, Inc. team, join and find out more here:
📍 https://luma.com/b8zggssl
See you next Wednesday, November 5, at the Antler’s Office 👋
What’s your opinion? Do you agree, disagree, or is there something I missed?
Enjoyed the article? The most sincere compliment is to share our work.
How can I help
Join the waitlist for our latest course on Agentic AI Engineering, where you will learn to design, build, evaluate, and deploy sophisticated, production-grade AI agents and workflows. Done with Decoding AI’s builder mentality, the course consists of 30+ lessons, with code and theory, during which you will build a real-world AI product that you can show off during your interviews. Done in partnership with Towards AI.
Images
If not otherwise stated, all images are created by the author.
Fantastic!!!! Will you stream it?
Good to know about the good 😊