Albert Wong
‘There Are Lots of Very Solid Engineers in the World:’ Albert Wong on the Outsized Role of Relationships, and Not Engineering Skills, at the U.S. Digital Service
| Albert Wong was an engineer and founding member of the U.S. Digital Service for more than two years, beginning in July 2014. Prior, he worked as a software engineer and engineering manager for over a decade at Google and Amazon. |
Albert Wong came to the U.S. Digital Service circuitously: He heard about the organization from another engineer, who heard about it from a teammate. He initially embedded for just a couple weeks, prior to the formal launch of USDS, but ended up having a much longer tenure with the organization.
Albert believed that successful digital service has less to do with technology and more to do with relationships. The thorniest problems often didn’t entail software bugs, but rather communication breakdowns between individuals, teams, and agencies.
Below, Albert discusses USDS’s evolution, the perils of over-competence, and why failures in public service are especially hard to reconcile.
June 3, 2024
Emily Tavoulareas:
Albert, tell me about your journey to the U.S. Digital Service.
Albert Wong:
It was a bit circuitous. I was at Amazon first, and then Google. Mikey Dickerson came back to Google and gave a talk about his HealthCare.gov experience. He made a statement that seemed ludicrous: “People just don’t show up to government. All we need are people to come in literally for a few weeks. That would be better than nothing.” My reaction was, “That’s ridiculous. There’s no way work could be done in just a few weeks.”
Brian Lefler heard the same talk and said, “I’m doing it. But I don’t want to go alone, so I’m going to ask Hannah.” Hannah was Brian’s teammate. She also happened to be my wife. Hannah said, “I can’t. Go ask my husband,” expecting me to also say no. But I said, “Sure, why not?”
Brian and I had questions: If we go, what are we doing? Are we going to be wasting anyone’s time? We’re just regular engineers — what can we do?
We signed up for a two-week stint. It was August 2014, and the team had just secured Jackson Place. Walking in, I said to Brian, “We have nine days. If we calculate working hours and times, that means every hour we spend is one percent of the whole project.” Eventually they take us in for a nine-day discovery sprint at the Department of Homeland Security (DHS). We were there with Charles Worthington, Mollie Ruskin, and Amélie Koran, plotting out our team and figuring out what was happening.
Emily:
What was the path from the DHS sprint to the Department of Veterans Affairs (VA) work?
Albert:
It was just supposed to be nine days. I was skeptical we would have any impact at all, but I tried really hard. When we had our first break on a Saturday I walked through the Smithsonian but was unable to read anything or speak because I was so burned out. That’s how hard we were going.
We saw there were project leads who were trying to do the right thing but didn’t have access to basic information. Meanwhile, there were people in their teams who were trying to surface that information, but were not able to do it. If and when we connected them together, they started making things work. It wasn’t that we knew anything special. We were both very solid generalist engineers with experience using lots of different tools, but there are lots of engineers like us in the world. What we were able to do in this weird setup was basically say, “You should trust this person and that person and they will be able to fix the problem.” It was knowing where to get real information, knowing how to jump the layers of bureaucracy, and being willing to ask questions.
I remember a product lead from the contracting side talking to a room of about 50 people. They said, “It’s going to take us a month to write a bash script.” And the government leads in the room didn’t know what that meant. I thought, “I don’t think this is right, why isn’t anyone calling B.S. on it?” So I asked why it would take more than a day to write a bash script. After the meeting, one of the team commented how the whole tone of the room changed once someone started asking questions.
Emily:
So it was a fundamentally different way of working: Who’s at the table? Who are you communicating with? How are you communicating? You basically created a cross-functional team and spaces for cross-functional communication.
Albert:
Yeah. Brian and I thought, “Okay, we’ve assessed this. But how do we actually create change that will last beyond when we leave?” And the only thing we could come up with was to create connections and help pierce organizational layers.
The problem I saw was more up and down than right and left. They weren’t able to deliver anything, as opposed to delivering the right thing. It was: “Don’t talk through your proxy to your proxy to your proxy. Why are you not talking to the decision maker?”
Emily:
It sounds like you were pushing back against the game of hierarchical telephone that happens in government. Where the fancy person in the suit has to report up to the fancier person in the fancier suit.
Albert:
Exactly. And there were different rules about how to bridge those hierarchical gaps. Back in the private sector, the default would be to create an after-work mingler where everyone had dinner together to build some relations. The one organizing would play host and pay for the meal.
We tried to do this on the final day thinking. “There’s nothing wrong. We’re just going to meet each other.” What we found out was that even though meeting up was okay, everyone was like “You can’t pay for this.” Instead, paying had to be divided per person down the penny. That was sort of a shock. The engagement was full of these and we had to adjust to differences in working in the public sector.
Coming back to Google, I could see how one can have an outsized impact in this setup in the short time period. And how there’s a lack of resources and no professional network integration. The White House got me through Mikey giving a tech talk, which Brian heard, who talked to my wife, who talked to me. Their networks are so separated. One of the biggest values I realized of doing this was to just create even one extra bridge between the networks as previously that had be nearly 0.
Emily:
That’s a really interesting way to think about the talent issues and challenges. When you have these networks that are so separate from each other, you need someone who can create that bridge to break through. And then once there’s a hole in the wall, people start mixing up and meeting each other.
Albert:
I was always remote, so I had a hard time fully understanding. I never knew how much of that approach we brought in from our side, and how much other people already knew to do it, but were having problems getting the authorization. There was always a strange feeling whenever you crossed the contractor-fed divide. People felt very worried about it.
The need to make connections between people was the bread and butter of USDS. People who joined had an intuitive understanding of how professional networks and communication worked. I can’t think of a single person in the team that didn’t have some feel for that. You had to think very carefully: “How are people going to interpret the meeting? How are people who didn’t get invited going to interpret it?” Everyone knew that was the battle. We were breaking the chain of command, and that has pluses and minuses.
It was something I thought an exceptionally large amount about because, being remote, I had to work extra hard to force and maintain relationships. On the other hand it also meant I was very cognizant of team dynamics.
Emily:
What do you think the minuses are?
Albert:
Anytime you go to an organization and make any change — even if it’s a good change — then like surgery, you are going to cause some amount of damage. If you disrupt the chain of command, at minimal, you’re going to break people’s sense of stability, harm their emotional health.
You can also cause confusion because people don’t know who’s actually in charge. Not that people ever knew that anyways — but there was a vague feeling about who was in charge. “I don’t actually know what their name is, but they’re probably in the org above here.”
So if you are going to do this, you have to be clear what outcome you wish to achieve to make taking such risk with other people’s experiences worth it.
Emily:
What do you think USDS did well?
Albert:
One of them is relationship building, or learning to leverage relationships when you don’t have authority, only have influence. You’re not anteing up your own career, you’re anteing up someone else’s. So you have a certain responsibility to not hurt people. That weighed on me a lot and I think that actually made me more trustworthy.
That initial sprint probably had the most long-term impact. We had no structure when we dropped in; we didn’t even know who was friendly. Brian and I initially designed the engagement process: What are we going to assess? How do we do this? How do we write a report? That ended up solidifying into discovery sprints. I don’t want to take full credit for it, but I think we shaped the foundation.
In the end, I think we were able to prove that change can happen and create some bridges between private sector tech and public sector tech. Nowadays I can talk about government service and people will have some context for discussion. They might even know someone who had done a stint in gov. When I first came back, talking about this experience was describing living in a foreign land. I think USDS’s longest term impact is not in its direct work, but in the changes to attitudes and relations two to three degrees away.
During the pandemic, a bunch of USDS alumni came together to help work on PPE distribution. I’ve seen that in other crises. We showed what could be possible and that is the most important.
Emily:
You were one of the first people in the mix. How did USDS evolve over time from your perspective?
Albert:
No one quite knew what this was going to be or what its potential was. There was a whole lot of information and so many unknowns — they didn’t even know what questions to ask. What really struck me about the organization was how smart and talented everyone was, but also… how little experience folks had handling strong emotional disagreements in a professional setting. The personal/professional divide was at times non-existent.
There was a lot of emotionality. It was weird to see people with these connections, this power, and these skill sets and knowledge — some of the best people I’ve ever worked with — reacting in a way that frankly felt like college or high school. You had this bimodal behavior: really good networking and really good observational strategy. But also, “They weren’t nice to me at a party last night, so I’m going to be short with them in the meeting today.” There was a point where everyone thought the org was going to blow apart from internal turbulence and self-created drama. When we finally got over it, we were able to highlight the good aspects.
But even then, there were a lot of problems due to, frankly, too much leadership talent.
There’s a term I’ve been using for this. Government dysfunction in the White House isn’t incompetence. It’s actually over competence. If people are so good at leading but misaligned even by just one percent, it can shear the whole org apart. And that’s what I felt in USDS. If you have a bunch of high-powered exec types sitting in a room, they cannot help but off-gas disagreements/unease about strategic direction. This will create friction in teams. This I think was the largest expected problem: friendly fire.
Emily:
It sounds like there was a moment when this subsided to some degree. Can you say more?
Albert:
The offsite was a turning point. That’s why Mikey was pushing so hard for it at the time. He was worried about all this. We hired an outside consultant to help with it. While there was disagreement about what the off-site needed to be or do, it still had the needed effect. People had dehumanized each other over the annoyances they had during the preceding methods. Forcing people to be stuck together made people remember, “Oh yeah, this is the person on the other end of the email who I’m pissed at and here is what brought us together in the first place.” One lasting thing we got from the offsite was the “USDS Values.” These have become lasting mottos that have preserved common ground and identity even amongst alumni.
Emily:
And how did things develop from there?
Albert:
There were projects where it felt like we failed, and that was particularly hard. Because in digital service, or any sort of public work, you really, strongly care about something. And when you fail, you don’t fail at a job — you fail your mission, you fail yourself. That is not an organization issue as much as a population sample issue.
Emily:
Albert, what seems to weigh heavily in your mind about this experience is related to culture and people. Is there anything you wish you had known earlier?
Albert:
I wish I knew how much the world is run by people with passion, which also implies a certain level of volatility in how they interact. And it was the first time I’d been in a group where most everyone was younger than me. It was weird to have these younger people in the White House — and behaving like it, too. That’s not a denigration. It was just like, “Oh, you’ve never had to actually handle long term care for an aging grandmother. So of course you will tell me that I should come to this event and that it’s the most important thing in my life.” Frankly, the dynamic tension of drive and life constraints was probably healthy for everyone.
Emily:
Is there anything that you are especially proud of during your time at USDS?
Albert:
I feel like I supported a lot of my teammates. There were times I was not my best self and I’m still sorry about some of those interactions, but overall I always tried to lift people up. And I think I succeeded sometimes.
Emily:
You absolutely did. Albert, I love how much you highlighted the outsized role relationships played in the work.
Albert:
Good talking with you, Emily.
Emily:
Thank you again for your time.