← POLITICO Tech
POLITICO Tech
Why DOGE led one government tech official to quit
00:20:15
Anne Marshall resigned from her role as director of engineering and data science at the U.S. Digital Service last month. The decision followed staffing cuts by Elon Musk’s Department of Government Efficiency. “This is not the mission I came to serve,” Marshall wrote on LinkedIn at the time. On POLITICO Tech, Marshall tells host Steven Overly about her decision to leave and what she thinks everyone is missing about DOGE.
Loading summary
Transcript39 lines
- [00:01]
McAfee Advertisement
Your data is like gold to hackers. They'll sell it to the highest bidder. Are you protected? McAfee helps shield you blocking suspicious texts, malicious emails and fraudulent websites. McAfee Secure VPN lets you browse safely and its AI powered tech scam detector spots threats instantly. You'll also get up to $2 million of award winning antivirus and identity theft protection, all for just $39.99 for your first year. Visit mcafee.com/incancel anytime terms apply.
- [00:33]
Stephen Overlea
Hey, welcome to Politico tech. Today's Monday, March 3rd. I'm Stephen Overlea. Last week on the podcast I talked to a former government official who made the case for what Elon Musk's Department of Government Efficiency could have been as a political appointee. Though she left the government before the Trump administration came into office. So today I wanted to talk with someone who stayed and experienced the windstorm of Doge firsthand. That someone is Ann Marshall. In 2023, she joined the US Digital Service, an office that helps federal agencies better deliver government services using technology. And last month, Anne quit, explaining her decision in a LinkedIn post that got a lot of attention. And since then, others at USDS have followed her out the door. On the show today, Ann goes deeper into her decision to leave and what she thinks everyone is missing. Here's our conversation. Anne, welcome to Politico Tech.
- [01:51]
Ann Marshall
Great. Thanks for having me here.
- [01:53]
Stephen Overlea
So let's talk about you recently resigned from your role as Director of Engineering and Data Science at the U.S. digital Service a few weeks into the Department of Government Efficiency taking over. What led you to that decision?
- [02:06]
Ann Marshall
So one of the things that originally brought me to USDS was the ability at USDS to make a real difference in the lives of Americans, to make material difference into it. I have a long history and industry and then that when you're talking about a new role, they'll give you terms like, oh, if we succeed at this, we're going to improve profits by 0.02 basis points.
- [02:29]
Stephen Overlea
Right?
- [02:30]
Ann Marshall
And that's not actually very motivating. But when you're doing government work like USDS has heretofore, the reward is if we succeed, n million of people can keep Medicare. People's material lives are valued, the ROI is completely off the charts and we're not talking 02 basis points. We're talking millions of people having their lives improved. And that is the core motivation to be at usds. Now why did I resign? Because I felt like the decisions that were being made and the actions that were being made by Doge made it impossible for our organization to continue on that mission and to serve that mission.
- [03:11]
Stephen Overlea
And what were those decisions and changes that you felt made that mission no longer possible?
- [03:17]
Ann Marshall
The short answer is they fired the wrong people. They fired a bunch of people imprudently, and they fired the wrong people. But in order to understand that, you kind of have to understand how USDS works. And usds, what it traditionally has done is there will be a multidisciplinary team that goes into the agency to try to support them. We're usually called in. Some people joke that we're like vampires, that we can't come in unless you invite us.
- [03:45]
Stephen Overlea
That's funny.
- [03:46]
Ann Marshall
But we bring in a multidisciplinary team. We don't just bring in engineers. We bring in design professionals who really understand what user experience should be or could be and how to make it effective. We bring in product people who are really good for defining the problem and figuring out what victory looks like and what are the steps to get there. We bring in procurement professionals. And people who are outside of government don't necessarily understand the term procurement, but the thing to understand is they are the people who understand how government contracts are written. And our experts can help the agencies write contracts so they will get effective delivery of the thing they're trying to deliver in a reasonable way that is not likely to have cost overruns. So these people are key to effective government and effective government contracting. But in the firings, they fired one engineer and they fired half of design and product and procurement. Design and product and procurement are key to us being effective. Not just building a thing or helping an agency build a thing, but to make sure we're building the right thing and we're building something that is easy for users to use and actually serves their problems. We work on technical problems, but the problem is usually not the technology. The problem is the product definition space, the making sure we have good design. It's like all these other parts. We have to have that whole team. And when DOGE fired indiscriminately, without that understanding of what we do and why we need these people, it was clear that we could no longer serve that mission. Like right now, I do not see a path where classic USDS is going to be able to continue to serve the agency partners to deliver for American customers, because we can't. We don't have the team anymore.
- [05:45]
Stephen Overlea
I imagine those changes felt very abrupt. I mean, just looking at it from the outside, it seemed very abrupt.
- [05:51]
Ann Marshall
It was very abrupt. Like we as a director Normally I have been in layoff situations. In industry, there's some process where you go through, okay, we have to cut heads like, where can we do it, what's critical, what's not. There was none of that. We had no input. We had like, literally we spent days going back, calling our people that report to me and say, did you get fired?
- [06:17]
Stephen Overlea
So you didn't even know if your own reports were terminated?
- [06:20]
Ann Marshall
Right. We didn't see the list before or after. We had to rebuild that. And so it's a real sign of how those decisions were being made at such a high level that they did not know what they were doing. I mean, that they hit procurement, our procurement team is very excellent, but also very small. And that they hit that the way they did clearly shows they are not trying to get more effective government contracts. They're not trying to use money more effectively because these are the people that do it. And my personal take is they don't actually understand government contracting. And so they literally did not know what they were doing.
- [07:02]
Capella University Advertisement
Imagine what's possible when learning doesn't get in the way of life at Capella University. Our game changing flexpath learning format lets you set your own deadline so you can learn at a time and pace that works for you. It's an education you can tailor to your schedule. That means you don't have to put your life on hold to pursue your professional goals. Instead, enjoy learning your way and earn your degree without missing a beat. A different future is closer than you think with Capella University. Learn more at Capella. Edu.
- [07:32]
Stephen Overlea
You know, you've drawn some comparison to your time in private sector and you worked at Amazon for many years, which is not a company that's known for having a soft work culture. Necessarily. When people say that Doge is just bringing a businesslike approach to government, how do you respond to that?
- [07:50]
Ann Marshall
Okay, so admittedly my entire technical career, except for my early years was at Amazon and Amazon is its own culture. Amazon and I think most of tech is a fundamentally data driven culture where it's investigating, it's digging in and actually doing an analysis and then deciding where to aim. Doge did not do that. Like a data point is, I was the director of engineering and data science. There are two parts. When Doge did our 15 minute interviews on the day, they interviewed the data scientists with the same non technical questions and loop that they did the product and design people, not the technical path. And a lot of people don't understand what the role in data science is. The way I tend to Frame it is. If you don't have data science, you're playing Battleship. You're firing things and see if it hits. If you actually want to have your. To fire at something and have it hit, you want a data science to tell you where to aim. That is a key role of data science. And in a space as vast as government, like, they're. They're priceless. They're. They're worth their weight in gold.
- [09:05]
Stephen Overlea
Right?
- [09:06]
Ann Marshall
And Doge didn't understand that coming in when they did their firings, they spared the data scientists. So apparently they learned something.
- [09:13]
Stephen Overlea
But is there a version of Doge that you think could have worked?
- [09:18]
Ann Marshall
There is absolutely a version of Doge that could have worked and could have been tremendous. There are absolutely issues in government and in government tech USDs, over its history, has been engaged with many different agencies. I would suspect maybe all the agencies at one time or another, hundreds of projects. We go into these agencies and we don't see hundreds of different problems. We see like five to ten different problems. Like, the set of problems are. Are small. The people who are very seasoned who have done their two tours start to joke about, oh, it's a set number 73. That's the problem we have. The number of problems that are happening in government are actually small. And that. And they span agencies. Agencies have very different cultures across their span. But yet all of these agencies of different cultures are coming up against the same struggles. And that tells me the root cause is above. It's in the system, it's in the contracting structure, it's in the OMB rules about how things are structured. And there is an opportunity that a disruptive change like DOGE could have really made positive change. What that would have need to look like is actually do some data analysis, figure out those problems, what are the things that are causing those problems? And that was a huge opportunity that they have not looked at.
- [10:45]
Stephen Overlea
Is that something that has just not been done before? You know, some of the rhetoric around Doge and some of the explanations for it have been, you know, attempts at reform in government have happened before that, you know, you need to be radical if you really want to actually make change. How do you feel about that?
- [11:05]
Ann Marshall
I absolutely think there's a degree of radicalness to make change, but there's also throwing the whole card deck in the air and figuring out, well, maybe they'll land somewhere else, but blowing everything up. You blow up the good structure as well as the bad. And while there's huge opportunities, like so much opportunities in procurement law about how to change how that's structured. Throwing it in the air doesn't get you that when you put those cards back together, if you didn't change the way you shuffle, you're still going to get the same problems. And that's what USDS has learned over its history is going in and being a disruptive change and then leaving is not effective. We have a bunch of core values and one of them is create momentum. It's not just getting this project successful, it's getting this organization into a pattern of operating that the next project is successful and the next. So it's teaching them how to be better. Is a lot of what we do. We talk about capacity building and that's difficult because it's not as fun. It's way more fun and gratifying to throw everything in the air and blow it up than to actually mindfully grow and nurture and develop the right things. And so that's the missed opportunity because we do need breaking changes. We do need breaking change in some of our rules. I would argue we need more government technologists, not less because we're relying too much on contractors to bring that. That's an OMB rule. And so that's where there is an opportunity to make change that will change the mechanisms. But they're not looking at that.
- [12:49]
Stephen Overlea
Well, it reminds me you were saying earlier about the government having such a high return on investment, but you're not measuring that in financial terms the way you would at a business necessarily. Right. You're looking at impact on Americans, things like that. I wonder what you see as the impact of DOGE kind of longer term on our government, how it functions. And do you have concerns about whether it can come back from these sweeping cuts that are being made?
- [13:16]
Ann Marshall
I'm very worried about it coming back because one of the differences between industry and government is industry needs to serve a core enough population to make money. Government serves everybody. And you don't get to decide that. Some Americans, they all have rights, they all need to be served. You need to support the whole breadth of that by blowing everything up and say we blow everything up and we rebuild it. We're going to only start with the core. Like we're not going to hit the special cases, we're not going to hit the immigrants, the, the low income people. These are the people frequently forgotten. And by blowing it up, those are the hardest people to serve. That's most of the work is on the edges.
- [14:01]
Stephen Overlea
Right?
- [14:01]
Ann Marshall
Blowing it up, you're going back to the center. Those people who need support. The Most are the last ones to get served.
- [14:08]
Stephen Overlea
There was part of your LinkedIn post when you posted about your departure that caught my eye. You said, if you're looking to start something to serve the American people the way USDS has for 10 years, let's chat. I'm curious if you think that we'll see some civic tech startups spun up by people who are leaving government or were pushed out of government as a result of what DOGE is doing.
- [14:33]
Ann Marshall
I sincerely hope so, because this is what I see not only with USGS people leaving, but TTS people and other specialists across government who are pushed out. They very much want to continue to serve American people. And the question is, how do we do that? So like, I have a huge number of people who are like, if you start something, let me know. But the question is, how do we get started? And so those are the conversations I'm trying to have now with people about, like, what, what would that look like? What could we actually do from the outside? How can we serve these people who are no longer being served by the government? And it's a hard question, but I, I'm hoping there's an opportunity because there's so much energy and intent and goodwill and excellent people really wanting to do this that I try to be optimistic that we'll be able to get something pulled together.
- [15:27]
Stephen Overlea
Perhaps you could call it a silver lining if that does happen. But you know, that could be a really fascinating outcome of this. Stepping back, I wonder, like, what do you see as the impact of DOGE on the government? Existentially, does the government as we know it look different going forward and can it come back?
- [15:47]
Ann Marshall
You know, that's an interesting question because what I see being broken apart are the pieces of execution that the government does. But what I don't see broken apart is how the government does it. And so we're keeping all the same friction but losing the actual mechanisms that make things work. So I feel like it will be a first class project to make government turning something on and back off and back on again, like a computer.
- [16:22]
Stephen Overlea
Right.
- [16:23]
Ann Marshall
Is valid. It is not valid for a government. It's more like an IC engine that needs like an alternator and a pump. And if you just drain the battery, there's nothing your alternator can. It's not going to start. So the problem is we have broken down the feedback loops that made government work, but we haven't broken down the things that make it slow. It's going to be a challenge to get it restarted.
- [16:46]
Stephen Overlea
One last question for you. Ian, you know, I know you're talking to a lot of journalists these days. Everyone. You're probably talking to a lot of people these days who have a million questions about Doge and what's going on and what it was like on the inside. I wonder if there's anything you're not being asked or anything you think, you know, the media or those of us on the outside are missing about all this.
- [17:07]
Ann Marshall
That's an interesting question. The media is very focused, or at least the media, most of whom harass me quite regularly, are very focused on trying to understand what's going on, what's happening, what are we not seeing? But I don't see the media focusing as much on this has happened. This is the change. How do we move forward? And that's partially because we're still at the table flipping stage. But, like, how do we rebuild? What have we lost? Who do we focus on? Like, that question of who is not served? Like, who is no longer served? And is that stuff that we as Americans want to be true? That whole space has not been explored. Journalists in general don't understand procurement. Nobody does. Everybody thinks it's buying pencils. And I think procurement reform is the path to make government better. There are concepts around agile procurement and how procurement needs to be done. And that is a whole topic in and of itself. That is actually how you reform and get government efficiency and spend less. Nobody is looking that. Like, it's very wonky, it's very technical, it's very specialized government, and people just don't understand. And that is the thing that everybody is missing. And I say that partially because I did not understand it coming in. And then we have phenomenal procurement people, or we did, to be honest. They fired one of my favorites who had saved the government millions of dollars that know this stuff, and nobody's talking about it. Like, the whole point's over here and you missed it. So that's my pet peeve, because as an industry professional, I did not understand that until they taught it to me, and then it's just paradigm shift.
- [19:11]
Stephen Overlea
Anne, thank you so much for being here on Politico Tech.
- [19:15]
Ann Marshall
Thank you for having me. It's been good talking.
- [19:20]
Stephen Overlea
That's all for today's Politico Tech. If you enjoy Politico Tech, please subscribe. And for more tech news, subscribe to our newsletters, Digital Future Daily and Morning Tech. Our managing producer is Annie Reiss. I'm Stephen Overleigh. See you back here tomorrow.