Darren Chinen has built his career in the heart of Silicon Valley, beginning in data analytics and data warehousing. His career really diversified during his time at Apple in 2012, when he built a data science team and became the first adopters of Hadoop in global business intelligence. He later served as the principal data architect, and head of data science and engineering at GoPro. Currently, Darren is the Senior Director of Web, Big Data, and AI at Malwarebytes. He describes his job as building award-winning and world-class web, big data, and AI teams in high-growth environments.
Short on time? Here are three key takeaways from Darren’s interview:
- Ask your engineers to act like owners. They should learn to take risks but also understand the consequences. People think differently about cost and value when they feel invested in the outcome.
- Have your team think in terms of superlatives. Everything they do should be the ‘-est.’ The coolest, the biggest, the best. Engineers want the opportunity to do something great.
- Strive to develop world-class celebrity engineers. Invest in developing their skills and helping them build their personal brand. When they recognize that you're taking the time to make them successful, they'll achieve more and want to stay.
What follows is a long-form write up of the key topics we discussed in our interview.
Darren Chinen developed his approach to building engineering teams when he was working as an engineer. “After spending so much time as an individual contributor, as someone in the weeds, I had a lot of time to think about what I wanted from my manager.”
He was attracted to the unique mentality in Silicon Valley, the most expensive place in the world. Why?"Because Silicon Valley has a unique culture when it comes to cultivating what I call an intrapreneur. These are entrepreneurs working inside of a company, as an employee, but who operate like an owner and have a lot of autonomy. They measure themselves with a different measuring stick."
Act like an owner
A manager's goal should be to get their team invested. “If [an engineer] thinks of spending as spending their own money, how they look at the cost of things becomes much more complex.” Engineers begin to look at the long-term cost instead of immediate investment. They are also willing to take more risks if the reward is more significant.
“The mentality and thinking is so much more long-term, and deep and healthy for the business if you can get people to really act like an owner.”
This approach works, Darren points out, once engineers are established. To get to this point junior engineers need to serve as apprentices. "Situational leadership is important. People want proper guidance when they come out of school, and as they grow up in the process, they begin to learn. That’s our investment." After engineers gain the knowledge and confidence, they can begin to take risks and act like owners.
"Tech has started this revolution."
When engineers on Darren’s teams have taken ownership, projects have moved faster. He noticed that the traditional approach to engineering had been, “waterfall-based, let’s measure twice and cut once.” That no longer works because in today’s tech world, companies are putting out new releases every day. To keep up, engineers need to move fast and leaders need to allow trial and error.
Think in terms of superlatives
Darren always asks his engineers to create things that are the ‘-est.’ They should be writing the fastest services, the coolest interfaces, or the “most awesome-est” solution to a difficult problem. Give people the challenge of being exceptional.
“One of the things that people get wrong about engineers is they think we are robots, we are mechanical people. But engineers want the ability to do something great. You have to give engineers that ability.”
This change away from "proven and safe" is being driven by the move to the cloud. According to Darren, data applications in the past were designed for a twenty or thirty-year lifetime. He notes that "the move to the cloud has really been a game-changer. It has forced us to change the way that we think about our infrastructure."
He shared an example where a data analytics project he was working on was able to gather billions of records at a time. At the rate that the data was growing, the storage and compute resources for analytics would quickly grow to cost them six billion dollars a year with the tools they were currently using. So they had to refactor. The team needed to create a solution that was the ‘-est’ to deal with exponential growth.
Become a world-class celebrity engineer
This can be complex. Darren asks his engineers to take time building their internal and external brand. He developed this approach after realizing that “sometimes you can't afford to hire the best, but you can afford to develop them into world-class engineers.” By giving each individual professional security, they became more confident and contributed more.
“I want my engineers getting a call from a recruiter every single day of the week. It's validation that we hired the best.”
There's plenty of ways to build your brand: publish articles, speak at conferences, and share your successes with others. Darren discovered that when engineers feel financially secure, they'll take the appropriate risks to move technology forward. Though some may think this approach could lead to the loss of good people, the reality is that Darren's teams stay together longer. “Ironically because I'm doing that, helping them build their brand, they stay.” He used coaching a sports team as an analogy. Success as a coach can be measured in terms of win-loss ratio or in terms of how many notable athletes your program produces. Darren feels that managers should not measure things like up-time or bug count, as much as they should work to create renowned engineers.
Ask engineers to score requests with a ‘fuzzy factor’
Every team faces communication challenges. When scrum masters or project managers (PMs) interact with engineers, they're usually either too prescriptive (so there's no wiggle room) or too vague (and way too high-level). Too prescriptive doesn’t give engineers the freedom, and too vague does not provide enough information to do what's needed. That's why Darren invented the “fuzzy factor," a scoring system that engineers apply to requests that signal to PMs how the request is perceived. The idea is for PMs to tell engineers what they want, engineers will acknowledge this, and include a "fuzzy factor" in a nice way.
“Information needs to go from product to engineers so the fuzzy factor is really just a way to figure out how to get enough information to the engineer, so they can in turn [offer actionable insights]."
The score goes from 1 to 10. A 10 is ultra fuzzy and means there's not enough detail to know where to start. A 3 means that the engineer has everything they need, and the project scope and objective is clear. When the score goes below 3, this means that the flow of information has reversed. Engineers finally have a chance to "surprise and delight, which is the entrepreneurial spirit in them."
The fuzzy factor is an efficient and non-confrontational way for engineers to give feedback to project management, and a way for those managers to get a feel for how well they're doing when they assign tasks. If it's mostly 10s, this indicates that a team needs to change how they communicate. If it's all 3s and 4s, then PMs can assume they are giving an adequate level of detail and direction.
Explore solutions with plussing and specialized teams
There's an importance in creating the right balance between exploitation of existing technologies and exploring new tech, while still achieving the company’s business goals.
“Engineering is a balancing act between exploring the next innovation and exploiting the technology you have, and collecting money."
Darren relies on plussing to achieve this balance. Plussing is about taking existing solutions and making them better by building on what's working, and also trying to improve existing tools. He gives his engineers time to try new things and make improvements. At Malwarebytes, “most of our new architecture has come from plussing projects.”
What do you do when the size of a problem is too big for a plussing project? Darren used a military analogy. The military doesn't send tens of thousands of troops to assess a threat, they first send a smaller team of Special Ops to quickly assess the scope of the threat. He does the same thing with development projects — have a small focus team first explore new tech, develop POCs, or think about new features. Some teams may end up being permanent, while others are temporary.
Someone once asked Darren, “how do I get promoted to a senior manager?” He said to build an organization of managers and he'll be "forced to promote you."
"If you're constantly in the process of building up other people and making them shine, you will be pretty successful in terms of engineering management.”