She spent a career making intelligence usable under pressure. Now she does the same thing for enterprise AI - and gets it trusted.
A demo that works in a sandbox is not the achievement. The achievement is the morning a logistics planner, a hospital supply manager, or a federal analyst opens a tool and trusts it enough to make a real decision with it. That gap - between a model that runs and a system people actually rely on - is where most enterprise AI quietly dies. Christine Williams built a company around the part everyone else skips.
Williams is the co-founder and Co-CEO of Foxtrot Services, a Washington, D.C. firm that turns tangled enterprise data into governed, working software on Palantir Foundry and AIP. She runs it alongside fellow Palantir alumna Nicole Sanders. The pitch is unglamorous and exactly the point: take AI out of the lab, hand it to operators, and stay in the room until it sticks.
"Most teams can get something working," she says. "Very few can get it trusted, adopted, and scaled. We build for that handoff from day one." It is the kind of line that sounds modest until you notice how few companies can actually deliver on it.
Foxtrot grew from two people to more than sixty, and it did so the unfashionable way: profitable and bootstrapped, without outside money, until the numbers made the case on their own. In early 2026 the company reported 9.1x year-over-year revenue growth and roughly 4x growth in both team size and customer base over a single year. Only then did it raise.
The Series A was led by Spicewood Ventures, with Scare the Bear Capital participating. The capital is pointed at the boring, expensive work of scaling delivery: more engineers, more delivery leaders, sharper go-to-market, and repeatable workflow patterns so that the next deployment does not start from scratch. Foxtrot is a women-led, Foundry-native Palantir certified partner, and its team members embed directly inside customer implementation teams rather than tossing slide decks over a wall.
The work spans manufacturing, healthcare supply chains, logistics, fintech, insurance, and the federal sector - high-stakes environments where a wrong answer is not a rounding error. That breadth is not an accident. It is what happens when a founder's instinct for consequence, formed long before the AI boom, becomes a business model.
Williams did not arrive at enterprise AI from a coding bootcamp. She came through the parts of the working world where being wrong has weight. Her background runs through intelligence analysis, homeland security, research, international relations, and national security - the disciplines of turning messy, incomplete information into something a decision-maker can act on. The throughline from that to Foundry is short.
Along the way she worked at Accenture, Peraton Labs, Victory Six Advisors, and Palantir Technologies, where she developed the deep fluency in Foundry and AIP - and the relationships inside Palantir - that Foxtrot is built on. She studied at Columbia University. She also, somewhere in there, found time to author articles on social responsibility and entrepreneurship and to learn passable Spanish.
Before she scaled enterprise AI, Williams founded the Child Soldier Relief Foundation and ran it from 2008 to 2011. She sat on the board of the Portland Conservatory of Music in Maine and served with Veterans Count / Easterseals Maine. She is a member of Women In Defense. The range is the tell: a person equally at home in a music conservatory boardroom, a veterans' services committee, and a data pipeline review.
It is easy to read a leadership bio as a straight line. Hers is not one. The common thread is not an industry - it is a habit of showing up where the stakes are real and the work is unglamorous, then staying until it is done.