You are viewing content from a past/completed QCon Plus - November 2020


Building Tech at Presidential Scale

With a small and scrappy team, the Biden campaign endured a highly talented primary field, an onslaught of misinformation, and made an all-staff transition to fully remote at the onset of the global pandemic. The global-scale challenges the campaign faced needed to be backed by a world-class tech platform to meet the moment.

In this talk, CTO Dan Woods will discuss the unique challenges of building and running tech for a presidential cycle. He will highlight the digital duct tape that held the pieces together and the data flowing. Dan will discuss how ML was applied at foundational points to reduce operating costs and give the campaign answers quickly. Furthermore, he will talk about how tools with names like CouchPotato, Tattletale, Conductor, and Pencil became a force multiplier of presidential scale.

Main Takeaways

1 Hear how a small team built a successful system for the political campaign for the primaries this year.

2 Find out what were some of the challenges to overcome and the architectural choices made.

What is the work you're doing today?

I'm a distinguished engineer at Target, which is effectively the highest level engineering position in the company. My emphasis and focus are on infrastructure and operations. So that's effectively the organization that I work in within the company. And the focus at the moment is on building reliable, high scale, durable, distributed systems. Before I joined the campaign, this is also what I did. The work that I did before was essentially building a platform so that we could write and deliver software significantly faster. Essentially a deployment in your face, but also creating an environment that software was able to consume services at runtime that made it so that teams don't need to figure out all the moving parts. How to build microservices and distributed systems.

What are the goals for the talk?

We built what I consider to be the best tech and in politics, quite frankly. And we did it in an unprecedented timeframe. We built dozens and dozens and dozens of microservices and supporting systems and complex architectures to accelerate what we were doing on the campaign. And quite literally, to save tens of thousands of human hours in just about every vertical of what it was that we were doing. I think what we did was extremely impressive. And we did it with a very long team and we did it in a really fast period of time. And I think when we look back on it, especially as I was putting the talk together when I look back on it, it's amazing what we were able to accomplish. And ultimately, I think, played a really big role in contributing to a successful primary outcome at a minimum, and then hopefully a successful outcome here in the general as well.

Microservices in a campaign is a novel concept.

Yeah. And to expand on that a little bit, the goal was to use as many stable vendor products as is humanly possible in an area where we don't need automation and optimization. We didn't build it. But the reality is if you look back on the primary, we were facing twenty two extremely skilled and very capable and competent opponents in the primary. The field of Democratic technologists was spread very wide and very thin. And not just technologists, to be clear, the entire campaign ecosystem was spread across those 22 different campaigns. We didn't have a lot of people in various roles. But for a huge portion of the primary, we were considered the front runner and we had to operate like we were a very big campaign. When you compare what we did vs. Hillary's 2016 campaign, our footprint as an organization was a fraction of the size of what they had. We didn't have a lot of people working in compliance and accounting and all those other areas to do that kind of manual entry. So we had to build automation in different areas. We looked at the technology footprint very early on in the places that we had to tie together the vendor systems. We found areas where we were able to optimize by building a common set of services that was able to automate a common set of tasks effectively. And that's kind of how we ended up in that world. It was never the intention to go in and say, OK, we're going to build a robust API platform and microservices and all this kind of stuff, but naturally, that's where the world ended up for us.

Are there any specific things that you hope people will walk away?

We brought a unique skillset and mindset to a presidential campaign from a technology standpoint. I was very privileged to have been able to hire the caliber of technologists that I was able to hire. We had a lot of people with experience and in basically an enterprise, being able to build robust distributed systems that maybe we wouldn't have been able to find had it not been such a big and important election year. But because we were able to bring those people in and really world-class software engineers and IT people and folks who had an emphasis on cybersecurity and an understanding of how it would all fit together we were able to do these things that in every respect we should have been able to do on a presidential campaign. But we also reap the rewards, right? We built systems that save tens of millions of dollars, quite literally, and we built systems that make sure that we were putting cybersecurity at the forefront of everything that we did. We were a very small and very scrappy team for a really long time. We were able to grow the team a little bit. But this is not a 100 or 200 person technology organization building these things. For a big portion throughout the primary, we had five software developers, myself included. And we got a really meaningful impact for just about everything that we touched. That ultimately is what I'd like people to take away from the talk is we put everything we had into making sure that we were going to win. And at the end of the day, we built some really interesting architectures and, more importantly, those interesting architectures paid meaningful rewards for us.


Dan Woods

Distinguished Engineer @Target, previously CTO for @JoeBiden & @HillaryClinton

Dan Woods was CTO for Biden for President during the 2020 election. Previously he worked building tech for Hillary Clinton's 2016 presidential campaign and before that he was a senior software engineer at Netflix. Dan is an O'Reilly author, an avid public speaker, and technologist....

Read more

From the same track


Using DevEx to Accelerate GraphQL Federation Adoption @Netflix

Wednesday Nov 18 / 01:40PM EST

In this talk, we describe how we enabled a large organization’s fragmented API offerings to unify under a fully federated GraphQL architecture by focusing early on a great developer experience.Not only does this help reduce the learning curve associated with federated GraphQL, it also...

Paul Bakker

Senior Software Engineer @Netflix, Author of "Modular Cloud Apps with OSGi"

Kavitha Srinivasan

Senior Software Engineer @Netflix


The World Is On Fire And So Is Your Website

Wednesday Nov 18 / 12:00PM EST

Learn how MoveOn architects and scales a complex ecosystem of custom tools that power political organizing work like rapid response mobilizations, get out the vote programs, and data-driven campaigning... on a nonprofit budget! We'll discuss the modern attention economy, the world of...

Ann Lewis

Senior Advisor for Technology Delivery at the U.S. Small Business Administration


Panel: Modern Architecture Patterns

Wednesday Nov 18 / 02:30PM EST

The way in which we build software is changing ever more rapidly.  Working out what technologies and approaches to bet on is, correspondingly, becoming ever more difficult. For software architects this pace of change means continual research and experimentation to keep on top of this...

Luca Mezzalira

Vice President of Architecture @dazngroup

Kavitha Srinivasan

Senior Software Engineer @Netflix

Paul Bakker

Senior Software Engineer @Netflix, Author of "Modular Cloud Apps with OSGi"

Wes Reisz

Platform Architect @VMware & Creator/Co-host of #TheInfoQPodcast, previous VP of Technology @Section

Colin Breck

Sr. Staff Software Engineer @Tesla

View full Schedule

Less than


weeks until QCon Plus May 2022

Level-up on the emerging software trends and practices you need to know about.

Deep-dive with world-class software leaders at QCon Plus (Nov 1-12, 2021).

Save your spot for $549 before February 7th