What I Wish I Knew When I Managed My First Engineering Team
April 20, 2021

Dan Pupius

As a developer at Google in the early 2000s, I often felt out of place. My focus at the time was Javascript, which many Googlers didn't consider "real engineering." But by the time I left Google, web programming was mainstream and through my contributions to Gmail and the frontend infrastructure used by dozens of products, I'd achieved the level of Staff Engineer. More importantly, I finally considered myself a pretty good engineer.

I moved on to Medium as an individual contributor. I valued teamwork and leadership from all angles, but didn't have any official management experience. In fact, I wasn't particularly excited by the prospect of being a manager, always feeling more comfortable leading from behind than in front.

While I bristled at "manager culture" — with its command and control origins — I also felt that the "no manager" movement had created a toxic void that left employees feeling unsupported or jockeying for power. So, when the day came that I was asked to step up and take lead of the engineering team, I decided to pursue a different path, that of the servant leader.

Luckily Medium was well primed for this approach, with both Ev and Biz, who had seen hyper-growth at Blogger, Google, and Twitter — actively wanting to do things differently this time around.

I saw my role as an organizational problem solver; instead of telling individuals or teams what to do, I worked to put in place support structures for junior engineers, identified owners for infrastructure, refined our hiring process, and generally worked to clarify roles and responsibilities. The goal was to empower individuals and set teams up to work in ways that fit them best, while also streamlining processes to reduce stress and confusion.

As we grew the team and evolved our processes, I questioned the status quo and evaluated processes from first principles. Here are some of the things I learned along the way.

Build a 10x Team, Not 10x Individuals

Team performance is about so much more than just the sum of the individuals involved.

A high-performing team can truly be 10x, overachieving and reaching goals ahead of deadlines. At the same time, a team of disconnected but exceptional individuals can perform terribly, if the conditions for collaboration are not there. It's imperative that managers create a collaborative environment, or every team member — especially in this remote world — will feel like an island.

Rethink Performance Reviews

Traditional performance review processes do not do a good job of serving individuals' needs. They tend to be backward-looking performance assessment, as opposed to forward-looking feedback — feedforward — that re-establishes expectations for the future and lays out a path for how individuals can progress, both in terms of the work they do and how they fit into the team.

Furthermore, coupling feedback with assessment muddies the water — who is the audience? The manager or the individual? I found that decoupling career feedback from performance assessment provides a more meaningful experience for individuals which actually helps them grow much faster.

Establish Rituals and Routines

Humans are ritualistic creatures. Building a drumbeat or rhythm into your work processes makes everything easier. With a good cadence, everyone on the team knows what's happening and when. In chaotic environments, teams become disjointed and find it hard to settle into deep flow states where the work seems to flow effortlessly. Also, not knowing what's coming day to day adds a layer of anxiety and stress as team members prepare for the unknown.

Constantly Observe, Diagnose, and Evolve

Keeping dev teams effective is a constant process of observing and diagnosing. Structure and practices will never be stable for long, so you need to build and exercise a muscle for change. Reorganizations are traumatic and disruptive, so aim for continuous evolution instead of "ripping the band-aid off" every so often, which will waste time as teams settle back in and adjust to new work parameters.

Always Remind Team Members About the Big Picture

People need to be constantly reminded of the vision and how their work fits into the big picture. If not, even the simplest tasks can feel pointless. Even if you feel like you are repeating yourself and becoming boring, it's important to constantly work to keep everyone aligned and focused on the North Star. Without that guide, everyone builds their own rowboat versus constructing a stable ship.

Dan Pupius is Co-Founder and CEO of Range
Share this

Industry News

May 06, 2021

Splunk announced the new Splunk Observability Cloud, the full-stack, analytics-powered and enterprise-grade Observability solution.

May 06, 2021

Gluware unveiled its DevOps for NetOps framework featuring Gluware Lab, its integrated development environment (IDE).

May 06, 2021

Ambassador Labs announced the new Ambassador Developer Control Plane (DCP), whichgives developers the ability to manage the entire modern software development lifecycle for Kubernetes environments using tools and processes that are familiar to them.

May 06, 2021

Code Dx and Secure Code Warrior have teamed up to launch Project Better Code, an initiative to tackle a major challenge facing innovative organizations today – pushing the pace of software development without compromising software security.

May 06, 2021

Pegasystems announced the latest evolution of its Pega Infinity software suite to help speed and simplify digital transformation (DT) initiatives, Pega Infinity version 8.6.

May 06, 2021

Accurics announced that its open source project Terrascan, which enables teams to detect compliance and security violations across Infrastructure as Code (IaC), now integrates with the Argo Project.

May 05, 2021

Amazon Web Services announced the general availability of Amazon DevOps Guru, a fully managed operations service that uses machine learning to make it easier for developers to improve application availability by automatically detecting operational issues and recommending specific actions for remediation.

May 05, 2021

SmartBear has added API testing support for the popular, open source event streaming platform, Apache Kafka.

May 05, 2021

Red Hat unveiled its Developer Sandbox for Red Hat OpenShift, an OpenShift-based development environment designed to enable organizations to accelerate the path from code to production for Kubernetes-based applications.

May 05, 2021

DevOps Institute announced the lineup for SKILup Days in the second quarter of 2021.

May 05, 2021

Idera announced the acquisition of Xblend Software.

May 04, 2021

ThoughtSpot announced the launch of ThoughtSpot Everywhere.

May 04, 2021

Perforce Software announced the availability of virtual devices (Android emulators and iOS simulators) as part of the comprehensive device lab within Perfecto’s Intelligent Test Automation platform.

May 04, 2021

LogiGear announced the newest release of its flagship TestArchitect™ Enterprise product, TestArchitect Enterprise 9.0.

May 04, 2021

Rafay Systems announced new enhancements to its flagship Kubernetes Management Cloud (KMC).