Engineering Culture

Engineering Culture

Help the user build strong engineering culture using frameworks from 19 product leaders who have built high-performing engineering organizations.

How to Help

When the user asks for help with engineering culture:

1. Diagnose the current state - Understand their team size, practices, and pain points 2. Identify the bottleneck - Determine if the issue is DevEx, org structure, talent, or process 3. Design for the right outcomes - Focus on speed, stability, and developer satisfaction 4. Balance structure with autonomy - Help calibrate how much process is needed

Core Principles

DevOps is capabilities, not tools

Nicole Forsgren: "DevOps is not a toolchain you buy. Marketing teams labeled toolchains DevOps because they wanted your money. DevOps is a set of capabilities - technical capabilities, architectural capabilities, cultural capabilities, and lean management practices." Focus on integrated capabilities, not just implementing specific tools.

Conway's Law is powerful

Dhanji R. Prasanna: "Conway's Law can be really, really powerful. You ship your org structure. What you're organized as in terms of teams and collaborating groups matters a lot to what you build." Organizational structure directly dictates architecture and product quality.

Engineers are becoming architects

Scott Wu: "We've thought about Devin as really allowing engineers to go from bricklayer to architect. A lot of it is just getting to the point where you can do the high-level directing and specify things exactly how you want." The future of engineering involves shifting humans from manual coding to architectural oversight.

DevEx is the foundation

Nicole Forsgren: "Developer Experience is what it's like to build software, day to day. The friction they face, the workflows they have to go through. It's important because when DevEx is poor, everything else just isn't going to help." Flow state, cognitive load, and feedback loops are the three pillars.

Optimize for clock speed

Albert Cheng: "The product experience of Duolingo actually changes multiple times per day for each user. They care a lot about the clock speed of the company." High-performance engineering culture is built around extreme experimentation frequency.

Require all functions to touch code

Brandon Chu: "All project management was just in GitHub, just commenting on PRs, and even marketers in order to upload a blog post, you'd have to commit and deploy it." Build technical culture by requiring all functions to engage directly with engineering tools.

High talent density reduces process need

Michael Truell: "We've been able to get away with not a ton of process yet on the engineering front by hiring people who are really excellent, level-headed, people who are less focused on external validation, more focused on building something really great." Hire for disposition to reduce organizational overhead.

Pair programming is underutilized

Farhan Thawar: "Pair programming is the most underutilized management tool in engineering, bar none. The throughput limiter is not hands-on keyboard. It's where is the good elegant solution?" Use pairing for knowledge transfer and finding elegant solutions.

Questions to Help Users

• "What's the biggest friction point in your developers' daily workflow?"
• "How does your org structure map to your product architecture?"
• "How much time do engineers spend in flow state vs in meetings?"
• "What's your team's 'clock speed' - how often can you ship and test changes?"
• "Do non-engineering roles understand the deployment workflow?"

Common Mistakes to Flag

• Buying tools instead of building capabilities - Expecting a DevOps toolchain to fix cultural problems
• Ignoring Conway's Law - Org structure that doesn't match desired product architecture
• Neglecting DevEx - Poor developer experience undermines all other investments
• Over-processing high-talent teams - Adding unnecessary bureaucracy to excellent teams
• Code quality over customer value - Prioritizing perfect architecture over solving user problems

Deep Dive

For all 23 insights from 19 guests, see references/guest-insights.md

Related Skills

• Technical Roadmaps
• Managing Tech Debt
• Platform & Infrastructure
• Design Engineering

Engineering Culture - All Guest Insights

19 guests, 23 mentions

---

Albert Cheng

Albert Cheng

"The product experience of Duolingo actually changes multiple times per day for each user... they care a lot about the clock speed of the company... they were really, really tight about it."

Insight: A high-performance engineering culture can be built around extreme experimentation frequency and tight, consistent processes.

Tactical advice:

• Optimize for 'clock speed'—the frequency with which the product can be updated and tested.

Timestamp: 00:35:46

Brandon Chu

Brandon Chu

"When I joined, all project management was just in GitHub, just commenting on poll requests and even marketers in order to augment or upload a blog post, you'd have to commit and deploy it. So there was no breaks given for folks that didn't want to touch code and stuff like that."

Insight: A strong technical culture is built by requiring all functions to engage directly with the technical tools and workflows of the engineering team.

Tactical advice:

• Use GitHub for project management to keep PMs close to the code
• Require non-technical roles to understand basic deployment workflows

Timestamp: 09:20

Chip Huyen

Chip Huyen

"CS is about system thinking, using coding to solve actual problem and problem solving will never go away because what AI can automate more stuff. The problem is just get bigger."

Insight: As AI automates syntax and basic coding, the core value of engineering shifts to 'system thinking' and holistic problem solving.

Tactical advice:

• Restructure engineering teams so senior engineers focus on peer review and architectural guidelines
• Train engineers to look at how different components interact rather than just local code fixes

Timestamp: 00:52:27

Dhanji R. Prasanna

Dhanji R. Prasanna

"I learned the most throughout this process is that Conway's Law can be really, really powerful. So it's the law that basically says you ship your org structure. So what you're organized as in terms of teams, in terms of collaborating groups and your operating model matters a lot to what you build."

Insight: Organizational structure directly dictates the architecture and quality of the products being built.

Tactical advice:

• Align technical strategy across silos to ensure teams speak the same language
• Standardize leveling and policies (e.g., what a senior engineer means) across the entire organization

Timestamp: 00:12:13

---

"A lot of engineers think that code quality is important to building a successful product. The two have nothing to do with each other. ... It really has very little to do with how well it was architected because the flip side of that Google video... supported more formats, it supported higher resolution... YouTube had none of this. It just had the one or two minute quick video thing and it's far and away, blown away its competition."

Insight: Product success is driven by solving user problems, not by the underlying architectural elegance or code quality.

Tactical advice:

• Prioritize solving the core customer problem over refactoring or perfect architecture
• Be willing to accept 'controlled chaos' and technical messiness if it enables rapid experimentation

Timestamp: 01:02:02

Gergely

Gergely

"I'm typically interested in the stuff that you might not read about in the traditional media, but us techies really care about... I wrote one about Facebook, one about Amazon, and I tried to talk with mostly software engineers to get a sense of how these companies work from a software engineer and engineer manager perspective."

Insight: True engineering culture is best understood by analyzing the lived experiences and internal processes of individual contributors rather than official company narratives.

Tactical advice:

• Conduct qualitative interviews with engineers across different levels to map out the reality of a company's technical culture.
• Look for 'scoops' or internal changes (like shifts in agile processes) to identify cultural trends.

Timestamp: 01:13:51

Gaurav Misra

Gaurav Misra

"I eventually joined the design team... a big part of what I did there was create this function called design engineering... The person who can think of the UX design it and also build it and launch it."

Insight: Creating hybrid roles like 'Design Engineering' allows for faster innovation by merging craft and reducing handoff friction between functions.

Tactical advice:

• Encourage engineers to learn design and designers to learn engineering to increase individual autonomy.
• Create a dedicated prototyping team to test high-risk ideas quickly within the main app.
• Hire 'full-stack' product builders who can own a feature from UX design through to implementation.

Timestamp: 00:43:51

Guillermo Rauch

Guillermo Rauch

"We have a ritual every Friday, we had it this morning, called Demo Fridays. And so it's very important to create the space for people to step out of that comfort zone and use AI. So us giving permission to people to build and ship things is part of that cultural backdrop that makes these things possible."

Insight: Foster a culture of shipping by creating recurring, safe spaces for employees to demo experimental builds and new technologies.

Tactical advice:

• Implement 'Demo Fridays' to encourage cross-functional building.
• Give explicit 'permission' for non-engineers to use AI tools to build and ship prototypes.

Timestamp: 01:24:30

Heidi Helfand

Heidi Helfand

"Switching also helps build that knowledge redundancy in your company... have multiple owners of a system. So not only one person is that tower of knowledge that owns that one system... pairing and switching pairs and test-driven development... this kind of redundancy built safety into our systems."

Insight: Regularly rotating engineers between teams and pairs prevents knowledge silos and builds systemic resilience.

Tactical advice:

• Implement pair programming as a standard practice
• Rotate pairs regularly to ensure multiple people understand critical systems
• Use team switching as a tool for both employee fulfillment and technical redundancy

Timestamp: 00:43:11

Howie Liu

Howie Liu

"I really believe everyone could learn how to be a software engineer if they wanted to... Everyone can learn how to be a versatile kind of unicorn product engineer/designer hybrid in the AI-native era."

Insight: Foster a culture where technical skills are seen as malleable and accessible to all roles, reducing the 'dark art' mystique of engineering.

Tactical advice:

• Encourage non-engineers to use AI coding assistants (like Cursor) to build prototypes
• Promote a 'growth mindset' regarding technical skill acquisition

Timestamp: 01:26:39

Logan Kilpatrick

Logan Kilpatrick

"In a world where you're constrained by the amount of GPU capacity that you have as a researcher... each new researcher that you add is actually a net productivity loss for the research group unless that person is up-leveling everyone else in such a profound way."

Insight: In resource-constrained environments (like GPU scarcity), team size should be kept small to avoid productivity dilution.

Tactical advice:

• Evaluate new hires based on whether they 'up-level' the existing team's efficiency rather than just adding headcount.
• Recognize that adding people to a resource-constrained project can actually slow down experimentation.

Timestamp: 00:43:19

Michael Truell

Michael Truell

"We've been able to get away with not a ton of process yet on the engineering front... by hiring people who I think are really excellent... hiring people that are level-headed... people who are less focused on external validation, more focused on building something really great, more focused on doing really high quality work."

Insight: A high-talent density team can operate effectively with minimal process if the hiring bar focuses on intellectual honesty and intrinsic motivation.

Tactical advice:

• Prioritize hiring for 'disposition' (e.g., micro-pessimism, bluntness, and focus on quality) to reduce the need for heavy organizational overhead.
• Look for candidates with a mix of intellectual curiosity and experimentation.

Timestamp: 01:00:39

Mike Krieger

Mike Krieger

"The team that works in the most futuristic way is the Claude Code team. They're using Claude Code to build Claude Code in a very self-improving kind of way... they would do very line by line pull request reviews... And they've just realized Claude is generally right... so can you use a different Claude to review it and then do the human almost acceptance testing."

Insight: High-velocity AI engineering teams are shifting from line-by-line human code reviews to AI-driven reviews and human-led acceptance testing.

Tactical advice:

• Use AI models to review PRs generated by other AI models.
• Shift the human engineer's role toward high-level acceptance testing rather than syntax checking.

Timestamp: 00:15:47

Nicole Forsgren

Nicole Forsgren

"DevOps is not a toolchain you buy. Marketing teams labeled toolchains DevOps because they wanted your money. DevOps is a set of capabilities. They're technical capabilities. They're architectural capabilities. They're cultural capabilities. They are lean management practices that predict speed and stability."

Insight: Engineering excellence is a result of integrated technical, architectural, and cultural capabilities rather than just implementing specific tools.

Tactical advice:

• Focus on technical capabilities like automated testing and CI/CD.
• Implement architectural practices such as loosely coupled systems.
• Adopt lean management practices to drive speed and stability.

Timestamp: 00:26:34

---

"DevEx is developer experience. And when we think about developer experience, we're really talking about what it's like to build software, day to day, for a developer. So the friction that they face, the workflows that they have to go through, any support that they have. It's important because when DevEx is poor, everything else just isn't going to help."

Insight: Developer Experience (DevEx) is the foundational layer of engineering performance, focusing on daily workflows, friction, and support systems.

Tactical advice:

• Identify and remove friction in daily developer workflows to prevent burnout.
• Focus on reducing cognitive load so engineers have brain space for innovative problem-solving.

Timestamp: 00:07:00

---

"One way to talk about it is kind of three key things that have components that are important of themselves, and they also kind of reinforce each other. Flow state is one of them, cognitive load is another, and then feedback loops are another."

Insight: The three pillars of a high-performing developer experience are flow state, manageable cognitive load, and fast feedback loops.

Tactical advice:

• Protect 'deep work' blocks (ideally 4 hours) to allow engineers to enter flow states.
• Minimize interruptions that break cognitive focus during the coding process.

Timestamp: 00:09:07

Ryan Singer

Ryan Singer

"Every designer codes. And I don’t just mean HTML. I mean, like running the app locally, going in to the place where that view is rendered to make that thing look the way that they want it to look... I mean, really codes, every designer."

Insight: A high-performing engineering culture can be fostered by blurring the lines between design and implementation.

Tactical advice:

• Encourage designers to work directly in the codebase to reduce handoff friction.
• Ensure designers can run the application locally to verify their work in a real environment.

Timestamp: 01:29:05

Sander Schulhoff

Sander Schulhoff 2.0

"It's really worth having an AI researcher, AI security researcher on your team... having that research background really helps. So I definitely recommend having an AI security researcher or someone very, very familiar and who understands AI on your team."

Insight: Effective AI security requires a specialized blend of classical cybersecurity and deep AI research knowledge that standard engineering teams may lack.

Tactical advice:

• Hire or consult with dedicated AI security researchers rather than relying solely on classical security practitioners

Timestamp: 00:51:04

Tom Conrad

Tom Conrad

"It was certainly the case that I was welcome to weigh in on the game design or the marketing... but my job was to build the software on time with high quality, period. And if I did those things, I'd be rewarded. And if I didn't, I would not be successful."

Insight: A culture with clear 'swim lanes' that rewards deep functional excellence can be more effective for shipping than a culture of generalist 'gadflies.'

Tactical advice:

• Define clear accountability for functional deliverables
• Encourage deep mastery of a specific craft rather than broad, shallow involvement

Timestamp: 00:38:11

Yuhki Yamashata

Yuhki Yamashata

"It's so important that you're using your own products. And I think we're in a very lucky position where all of us can get creative around using Figma in some way... one of the first things I did when I arrived was we were a little bit more of a memo culture, and I was like, you know what? We should be a deck culture because we can build those decks in Figma, and just that act alone allows you to encounter a lot of issues and for you to get familiar with it."

Insight: High product quality is driven by forcing the entire organization to use the product for daily tasks, creating personal accountability for bugs.

Tactical advice:

• Switch internal workflows (like memos to decks) to force usage of the product
• Encourage 'bottoms-up' bug fixing where engineers fix things they personally find embarrassing
• Use the product for non-standard internal processes like performance calibrations

Timestamp: 00:35:27

Farhan Thawar

Farhan Thawar

"Pair programming... is the most underutilized management tool in engineering, bar none. ... The throughput limiter is not hands-on keyboard. It's not like we're both sitting there and the limiter is like us trying to get through the keystrokes onto the screen. The limiter is where is the good elegant solution?"

Insight: Utilize pair programming as a primary tool for knowledge transfer, intensity, and finding elegant technical solutions.

Tactical advice:

• Use tools like Tuple for high-fidelity remote pair programming
• Encourage pairing during incidents or for complex 'wayfinding' tasks
• Delete and rewrite code immediately after pairing to find a more elegant second-pass solution

Timestamp: 00:22:30

Scott Wu

Scott Wu

"One of the ways that we've kind of thought about Devin in building Devin is really allowing engineers to go from bricklayer to architect, so to speak. A lot of it is just getting to the point where you can do the high-level directing and you can basically specify things exactly how you want."

Insight: The future of engineering culture involves shifting the human role from manual coding to high-level architectural oversight and specification.

Tactical advice:

• Encourage engineers to focus on system design and problem specification rather than boilerplate implementation
• Adopt asynchronous workflows where humans direct multiple AI agents in parallel

Timestamp: 00:24:48

---

"Our whole team is only like 15 engineers a year. We use a ton of Devin when we're building Devin. Most folks on the team are definitely working with up to five Devins at once, and so Devin merges like several hundred pull requests into production in the Devin code bases every month."

Insight: A high-leverage engineering culture uses AI to allow a small team to produce the output of a much larger organization.

Tactical advice:

• Measure team productivity not just by human output, but by the ratio of AI-driven pull requests
• Train engineers to manage multiple concurrent workstreams using AI agents

Timestamp: 00:00:00