About FNS

Who we are, where the work comes from, and what we believe.

The short version

FNS is an AI for Good practice founded by Colin Lyons in 2023. We design and build responsible AI for programs that connect communities to critical resources: benefits, rebates, health services, and the organizations that deliver them.

The longer version

I came up through operations. IT support in a 24/7 hospital system, where downtime meant a nurse couldn't pull up a chart. Then lead technical operations engineer at a $2 billion fintech, where I ran incident response and built the internal tooling that let operations, customer service, and compliance teams move faster: payment processing, refunds, identity verification. Years of watching small teams drown in work that mattered, and building the systems that kept them above water.

That is exactly the job I do now. The tools changed. The job didn't.

Before any of that, the through-line was already there: data analytics for Feeding America's Native American outreach, a statewide listening tour across all 64 Louisiana parishes, health screenings that connected thousands of families to care, a multilingual campaign putting the California Earned Income Tax Credit in front of the families it was created for.

I founded FNS because the programs doing the most important work tend to have the thinnest operations. A statewide program run by a handful of people. A nonprofit where the executive director is also the grant writer and the IT department. AI, built responsibly, is the first technology that can genuinely widen what a small team can hold. That is the work I want to spend my career on. This is not a market opportunity I am chasing. It is the work that lines up what I am good at, what the world needs, and what I care about.

What we believe

AI is a tool that makes teams more capable. It never replaces people, and we won't build systems designed to.

Humans own every consequential decision. Always.

If a number isn't traceable to a source, it doesn't get published. That goes for our clients' materials and for this website.

Honest answers beat sold answers. If AI is the wrong tool for your problem, we'll say so.

The work says it better than the bio.