Requiem for 18F

A small government unit did what DOGE is pretending to do

From Don: On midnight last Friday, the federal government tech unit 18F was eliminated. Much of what it did centered on improving federal tech services and contracts to generate better outcomes at lower costs. It was also committed to openness and transparency, making their code and guides open access. In other words, 18F did not just work on products: they worked on frameworks for making tech work for government, frameworks that other parts of government and other governments could use. While all that knowledge has been erased from federal websites, former 18F officials have saved it here.

18F employees, from deleted 18F website

Most members of the public have little reason to be aware of 18F, or will understand what its loss signifies. So this post tries to capture that, by including the voices of those who worked there. More broadly, we are in a moment of civic education, when functions or offices we did not know existed are being eliminated or downsized. Those doing the cutting say those public employees provided no value, so the cuts are a savings. I am asking you to also listen to the stories of those whose work is being eliminated, and judge for yourselves.

First off, Waldo Jaquith, a software developer and procurement professional who worked in leadership roles at the GSA (18F’s parent agency) and Treasury during the Biden administration shared some insights in his personal capacity:

For 11 years, 18F was a cost-recoverable division of the General Services Administration, charging agencies for their expertise, acting as consultants internal to government. Its cost to government was negligible, but its benefits were huge.

I worked at 18F from 2016–2020. We brought small, cross-functional teams to work with federal, state, and local agencies to solve discrete, urgent technology problems. The bill was cheap compared to the big consultants, while the business model was optimized for irrelevance—empower the agency partner, transfer knowledge, build up the agency team to the point where they no longer need 18F’s help.

18F did two things, both for agencies that hired them to help with projects: it built software and it taught agencies how to hire & oversee vendors to build software. The former raised the bar by showing agencies what “good” looks like, the latter allowed those practices to expand sustainably.

18F was a non-conflicted partner for local, state, and federal government that taught them how to build software, how to procure it, maybe held their hand through the procurement process, and then left, knowing that the agency didn’t need 18F’s help anymore. That was literally unique, and now it’s gone. I do this same work at a non-profit today, and we work hard to ensure that our partners trust us. The magic of getting that support from 18F was that the trust was assumed—it was obvious that a federal agency wasn’t going to try to upsell you, that they didn’t get a kickback for steering you to Oracle, that they weren’t going to lie to you.

Once, I was part of a team of four people at 18F that saved the Department of Defense $500 million with a single, three-day project. 18F projects saved many millions of dollars as a matter of course, while delivering better results for Americans. Another time I was part of a two-person team (our billing worked out to about half an FTE, as I recall) that partnered with a court for 18 months to help them build, from scratch, an open source case management system to drive the operations of the entire court. The whole thing cost them just a few million bucks, vastly less than they’d been spending on their old system, and today their cost of operating it is a rounding error compared to their old infrastructure costs. All the work 18F did for all their agency partners was open source. Public money should produce public software, for public inspection.

In short, 18F works just how Musk and team pretend that they want government to work. But when his team found it, they destroyed it. 18F’s work is evidence that government works well, which undermines their message that it doesn’t. 18F’s parent agency, the U.S. General Services Administration, turns a profit as an agency. So it has to be destroyed too.

To anybody in leadership at the state or municipal level: 18F’s destruction makes this the perfect time to hire experienced technologists, which states all need very badly. Most 18Fers would love to stay in public service. They are spread throughout the country.

The work that I led at 18F I naturally feel was very important (I hope all 18Fers felt the same way about their work): codifying the procurement principals that we’d all identified there over the years. I thought it would have a tiny audience, specifically the legislative fiscal directors of every state legislative chamber in the country. Instead it was a surprise hit, being downloaded over 100,000 times in the month after its release. In the years since it has become a foundational document in the government software space.

The work I do today at small civic tech nonprofit is simply what I did at 18F—software procurement, budgeting, and oversight—except I’m a team of one, dependent on grant funding. I have sent many agencies to 18F when they’ve needed large-scale support. But I have nowhere to send them now. I wish we could hire a bunch of 18Fers, but we’re a small organization, reliant on grants for funding. We’ll work with our state and local partners to create positions appropriate for these folks, and help to make those matches—perhaps that will help to retain this talent in the public sector.

18F faced a lot of threats over the years. In the beginning it was mostly from within, frankly. It’s the way of digital services that they break a bunch of rules to get started. Then the threats were external. But I never thought a threat was that it was just too effective of an organization for Republican myth-making. That, in the end, was what killed it.

---

Don again: Why was 18F targeted? As Waldo and others have described, the 18F model was anti-ethical to DOGE in terms of its goals. But there is another possible explanation. Brian Merchant has noted that right wing media criticized 18F as a group of radicals, mostly by focusing on the fact that it was an inclusive workplace, especially for LGBTQ employees. Shortly after, Musk posted this:

The targeting of LGBTQ employees for purges, especially those who are trans, has been a theme of the Trump administration. About 100 intelligence officers were fired for discussing their identity on a workplace chat, and Trump is driving trans military officials out of service. How much did this shape the decision to get rid of 18F? DOGE spokesperson Katie Miller disparaged departing civic tech employees for having hung “trans flags” at their workplace. Nowhere is there a question about the quality of the work of these employees; the focus of DOGE is on worker identity, not their skills or experience.

I’m also going to share some excerpts from posts made publicly by others who worked with 18F. What comes across most consistently is the sense that 18F was doing the work that DOGE promised to do: making government work better using tech, achieving real savings while improving services. You will notice how much the 18Fers are focused on the work and its impact on the public, not identity.

You can start with this thread from Dan Hon on Bluesky, describing his experiences working with 18F.

Lindsey Young, the Executive Director of 18F and its longest running employee posted:

We, like our many allies, had the “radical” idea that the government should be responsive to the needs of real people. We assembled amazing teams of technology professionals from different specialities who could work together and learn from each other. We shared what we learned with everybody.

I saw, time and time again, where we stood up for partners who were getting taken advantage of by vendors, or just needed help turning a vision into reality. We could make a simple website or a complicated system, we would do what we needed to best serve the mission and the public. We didn’t upsell anyone, we tried to teach our partners how to do what we did. I see them still prospering years after working with us.

We have proven methods that could be replicated, so we helped even more people through guides and writing. Those people are still going. And I am cheering them on.

We were living proof that the talking points of this administration were false. Government services can be efficient. You can work with agencies as they are now and work with them to better manage their services.

This made us a target. People who own skyscrapers are afraid of 100 people who made websites better. Not because of the latest tech fad, but because we proved that the government can be fixed, the government can be made better and the government can work for the people.

Adam Kendall said:

You should know that with the elimination of 18F deep in the night Friday that DOGE and this administration is not serious about eliminating waste or fraud. The very reason 18F existed was to provide industry leading expertise to other parts of government to help prevent them from being continually ripped off by vendors who masked their bad behavior, oversold or even outright blew budgets and timelines to increase revenue off your dime. I am proud of my time with 18F and stand behind those that were cruelly let go without regard to the impact that such a move would make. 18F will live on.

Rachel Grossman a UX Researcher at Amazon, posted:

I'm sitting with the news of 18F. From my first weeks as a designer nearly a decade ago, they changed the way I think about the field, and my role in it: Design is a verb, not a noun, and the best work happens in the spaces between disciplines — an infinite game.

I "grew up" through their open source guides and methods. But more than handing me the tools, they indirectly taught me how to do the work; they taught me that facilitation and collaboration are the backbone of design — that this craft thrives when we break down walls, speak in plain language, and build bridges across roles and disciplines.

What 18F built lives on in how I practice today. From UX strategy to research to service design, I've returned countless times to their open resources. Each use reinforced a central truth over the years: design excellence isn't in the perfect interface — it's in the alignment of values that feeds what we build.

No doubt, I'm still learning how to live these values fully. But the 18F ethos remains my north star; my greatest impact comes not from perfection, but from openness, collaboration, and shared purpose.

I don’t know what happens next for the type of civic tech that 18F represented, but am choosing to remain hopeful even as we record what was lost. The conflict with the value they provided, and the alternative that Musk is offering, is only going to become greater over time.