Can a traditional, 175-year-old news organization operate with the same digital sophistication and engineering velocity as a Silicon Valley tech giant?
For years, internal debates at The New York Times centered around an identity crisis: Are we a tech company, or are we not?
Today, that question has been decisively put to rest. The reality is that delivering world-class digital applications for journalism, games, cooking, and product recommendations requires an engineering operation that mirrors the cutting edge of the technology sector.
Senior VP of Product Engineering Brian Hamman and Chief Technology Officer Jason Sobel sat down with us to explore how they conquered structural silos, eliminated undifferentiated heavy lifting, and designed an architecture capable of sustaining a world-class digital enterprise.
Brian and Jason joined us to discuss:
When an organization scales, its internal structural silos inevitably begin to manifest within its software architecture, a phenomenon known as Conway's Law.
At The New York Times, historical news desks and regional product divisions created a highly fragmented ecosystem.
Teams were entirely “full-stack,” meaning every group had to manage everything from baseline cloud infrastructure deployments up to the user-facing digital experience.
“There was broad awareness that we had a lot of tech fragmentation,” Jason explained. “We had duplicate systems solving the same problems... It’s not that differentiated. There’s probably not a lot of value in ten teams having seven different approaches to it.”
This structural fragmentation resulted in massive amounts of “shared toil” — undifferentiated heavy lifting that consumed technical cycles without directly moving business metrics forward.
To build an explicit case for change, Jason utilized qualitative developer surveys to show stakeholders that some teams were burning up major chunks of their time on purely infrastructure-related maintenance rather than feature development.
“I was able to show that to stakeholders, and say, ‘Look, this team over here that you want [to grow] the subscriber base or [make] cooking a lot better, they're only spending twenty percent of their time on those problems,’” Jason recounted. “‘They're spending eighty percent of their time on this technical work. We can take that eighty and make that fifty, forty, twenty, depending on the team.’”
The structural antidote to this fragmentation was the creation of the Delivery Engineering Shared Platform (DVSP).
The mandate was clear: consolidate common infrastructure tasks, such as spin-ups, deployments, and continuous integration pipelines, into a centralized, repeatable platform layer.
However, funding internal platforms requires a difficult, long-term stakeholder conversation.
“The story was, ‘We're going to do this, and it's going to take time from your roadmap, and it's not going to immediately pay off in any metric that you as a business leader pay attention to,’” Jason shared. “‘But it's going to give leverage to all our engineers so that going forward, they can do a lot more in that space.’”
Brian noted that while the team had attempted to build platform components in the past, executing this strategy successfully required a distinct cultural shift.
“Historically, there was a feeling like, ‘Are we a real engineering team? Do we really deserve a real platform? Is this hubris that we think we can build it?’” Brian said. “ We didn't have the full conviction or even necessarily the belief that we could build it and it would be good.”
Coming from deep Silicon Valley roots at Facebook and Airbnb, Jason acted as an external catalyst, giving the organization the validation and conviction it needed to execute a large-scale platform strategy.
“Having Jason come in, who has built this, who has seen this at other organizations, and be able to say, ‘What you guys are doing makes sense. This can work, and it can work in an organization this size,’ helped put us over the edge,” Brian recalled.
Today, the platform stands as a profound point of pride that accelerates development across the entire enterprise.
Centralization drives systemic efficiency, but it introduces a distinct organizational tension: if you standardize too rigidly, you risk choking off creative experimentation.
Innovation historically occurs in protected, unstandardized silos where teams have the freedom to move fast and break rules.
Brian views organizational architecture not as a fixed destination, but as an oscillating pendulum.
“A tension there and a challenge that we're always wrestling with is the more you break down the silos and centralize things, the more you can also stifle innovation,” Brian observed. “We're trying to be very thoughtful about how we can standardize [and] centralize to raise the water level for everybody, but do it in a way that doesn't create so much lock-in that people are going to try to escape.”
Jason echoed this perspective, emphasizing that leaders should embrace structural iteration rather than fearing it.
“It's hard to be perfect, and so we are probably going to overcorrect in one direction or the other,” Jason said. “Eventually, that's going to get significant enough that we're going to have to make some changes, so we start swinging back the other way.”
Looking toward future technical horizons, agentic AI frameworks and next-generation coding environments like Cursor and Claude are fundamentally expanding what a solo developer can build.
This shift is uniquely powerful at an institution like the Times, which Brian mentioned naturally attracts multidisciplinary, hybrid professionals who straddle the line between journalism, design, and software engineering.
By leveraging AI assistants to manage complex technical heavy lifting, a single creator can now bridge gaps that previously required full cross-functional engineering teams.
“What AI has unlocked for some of these people is basically [a single] person being able to take something from start to finish,” Brian explained. “If you can understand the whole problem, not just the technical parts of it, but the journalistic, the design, the business… For a small number of people, it really has unleashed them.”
Ultimately, matching digital products to the absolute pinnacle of editorial standards requires a cultural foundation of trust, extreme curiosity, and deep personal accountability.
For Brian, the ultimate blueprint for leading through systemic industry change can be distilled down to a single piece of advice received early in his 20-year career.
“If you see something that needs to be done and it's not anybody else's job, it's your job. Go do it, solve the problem, and if nobody tells you to stop, that's now your job.”
Craving more? You can find this interview and many more by subscribing to Evolving Industry on Apple Podcasts, on Spotify, or here.