Flatiron Health’s Unique Approach to SRE Could Save Lives — Literally

Tzvi Friedman once had a dream to be both a doctor and a tech professional. 

Despite not having a medical background, Friedman's work in the health tech space as an engineering leader at Flatiron Health has had a positive effect on the lives of many cancer patients. As a staff engineering manager, Friedman supports the success of site reliability engineers while striving toward the company’s mission to improve and extend lives by learning from the experience of every person with cancer.

Site Reliability Engineering, a Google-originated discipline that evolved during the early 2010s to serve the reliability needs of web-scale organizations, bridges the gap between operations engineering and software development. Given the rise of cloud-native infrastructure that often blurs the line between infrastructure and software roles, Friedman noted that partnership should be the primary goal of a site reliability engineer (SRE); a philosophy that Flatiron has embraced. “Our SREs sit with our developers on a day-to-day basis, both influencing and being influenced by other engineering teams,” Friedman said. “I believe that’s the future of site reliability engineering; in order to get a seat at the table where software engineers are making early decisions about their architectures and processes, we need to be building those relationships at the design phase.” 

By working close to developers, SREs avoid the pitfalls that many organizations find themselves in, gatekeeping progress behind inefficient manual processes necessary to ensure reliable production environments. With a more intimate knowledge of the product codebase, that reliability can be baked in from the beginning as developers and SREs collaborate at earlier phases of the development lifecycle. 

Given that the Flatiron’s various products have differing reliability needs, Friedman noted it is essential to have an site reliability engineering function that can be integrated  across various  engineering teams as opposed to one central site reliability engineering team. While this structure presents some challenges—such as ensuring alignment among decentralized SREs—Flatiron’s focus on collaboration makes it easy to overcome these roadblocks. By framing site reliability engineering as a cultural initiative, Friedman and fellow leaders have more room to emphasize partnership-building as a core element of success. 

Operating within an environment that encourages cross-functional teamwork doesn’t just improve reliability capabilities. According to Friedman, it also empowers SREs to build out their skill sets while working alongside passionate individuals pursuing a noble mission. 

Cultural Evolution

Friedman recognizes that site reliability engineering  is an incredibly diverse field; people come to the discipline from a wide variety of technical backgrounds and experiences. While many skills overlap, each SRE also possesses unique knowledge areas, which is why Friedman considers it essential to give them space to learn from each other. 

“This is a place where SREs can come together to share skill sets, problems or solutions and provide mentorship,” he said. The more we can make collaboration valuable, the more it will happen organically.”

This is achieved partly through a standing biweekly forum, during which time team members from across the company come together to discuss roadblocks and brainstorm solutions. Friedman believes efforts such as this can jumpstart the creation of crucial relationships. 

When SREs aren’t busy tackling a task, they can stretch their skills in other ways. Flatiron’s annual learning and development stipend can be put toward any type of professional enrichment, from online courses to books. 

According to Friedman, the company’s “thoughtful, employee-focused” performance review process also helps SREs steer their growth journeys. In his mind, performance reviews are a comprehensive conversation with engineers to help them get the feedback they need from peers and managers to succeed in a rapidly changing technical landscape. As a manager, Friedman sees the annual review process as a culmination of the work that a manager and engineer put into communication and feedback throughout the year.

“Nothing should be a surprise in a performance review. It is a documentation of achievement and statement of goals that can help frame growth and success in a way that shows thought and care.” said Friedman. “I put more work into performance reviews than I do into almost anything else.” 

“I put more work into performance reviews than I do into almost anything else.”

Friedman believes that his focus on team members’ growth and goals stems from the support he has received throughout his time at Flatiron. After spending some time in an engineering role, he was offered the chance to take on leadership responsibilities. “At many companies, new managers are thrust into their new role with very little support and a mandate to just make things work. Flatiron spends a lot of time and resources making sure new managers have the tools they need to thrive and build great teams.”

This level of support and facilitation is common across the company, as Friedman has seen countless employees move between departments. “There are members of my team that internally transferred because they wanted to try something new. Each one of them was encouraged to take on the new challenge and was given the support they needed to make the shift as seamless as possible.” In other words, Flatiron encourages its people to push their boundaries and embrace opportunities to be impactful in ways they may not have expected. 

“Flatiron invests a lot in its people and their professional growth,” he said. 

Friedman joined Flatiron in part because the company is cloud-native at its foundation and built from the ground up with infrastructure-as-code as a core engineering value. This means the company approaches technological decisions with a focus on automation and scalability. 

Their cloud infrastructure stack is based on Terraform and Ansible as the base building blocks, and they encourage innovative engineers to work with site reliability engineering in using these building blocks in supported golden paths, ensuring reliable deployments of their services. The company has a burgeoning community of open-source contributors. “We like to build upon the open-source contributions of the rest of the community, since that’s worked for us really well,” Friedman said. 

‘Deeper Than a Tech Stack’

For Friedman, Flatiron’s site reliability engineering culture goes deeper than a tech stack—it comes down to the people. 

He and his colleagues have bonded over the past several years, despite team members being spread across several time zones. He believes the team has helped maintain a collaborative, open, and empathetic environment by building a culture that goes beyond scheduling virtual happy hours. “It’s about changing the way we address people’s working styles and comfort levels—setting the stage for people to bring their best selves to the table,” he said. “Hybrid work is different for everyone. When we give people the ability to work in ways that make them comfortable, we’re setting them up for success.”

While the company provides ample opportunity for connectivity among its employees, Friedman believes that a workplace culture grounded in authenticity and compassion is the secret sauce. “My team stays connected because we’ve built a culture that values human connection. We celebrate our wins, we share our struggles, and we really do care about the human behind that zoom tile. Being remote doesn’t really change that.” 

Ultimately, Friedman believes that SREs can only thrive in a culture of psychological safety. “The key is giving people opportunities to grow and get out of their comfort zone while giving them room to experiment and sometimes fail,” Friedman said. “That strong sense of camaraderie allows us to experience the cycle of learning, failure, and overcoming challenges as a group. It makes the highs higher and the lows easier to get through, and so far I think it’s working out really well.”