Key Takeaways

  • Tech burnout jumped by 10% in a single year, climbing from 44% to 55% as teams adjust to AI-driven workflows.
  • Fiona Fong, head of engineering for Claude Code and Boris Cherny's manager, noted that software teams are shrinking from 5 to 10 engineers down to isolated solo operators.
  • The shift from writing code to prompting models turns builders into managers, replacing the joy of direct craft with supervision.
  • Molly Graham argues that leaders must openly acknowledge the grief of lost craft before expecting teams to embrace new workflows.

The Shift from Rowing to Steering

Software engineers and product builders are exhausted, but not just from long hours. They are burning out because the nature of their daily work changed overnight. Rachitsky shared survey data showing that burnout among tech workers spiked from 44% to 55% in a single year. That jump reflects an emotional crisis across the industry.

Graham captured the feeling with a simple metaphor about builder identity. “She said like the job used to be rowing and now it's like steering,” Graham explained. “And I was like, I feel like there's a lot of people out in the world right now that are like, I don't want to steer. Do you know what I mean? I want to row.”

For two decades, software craft meant sitting in a flow state, writing code, and solving hard technical puzzles with peers. Handing the keyboard to an AI model turns the engineer into an editor, a code reviewer, and a prompt manager. If you got into engineering because you loved the physical act of building, being forced to manage an automated tool feels draining rather than exciting.

The Loneliness of the Shrinking Team

Beyond the loss of craft, the social fabric of engineering is thinning out. On a recent podcast episode, Rachitsky spoke with Fiona Fong, who leads the engineering group behind Claude Code at Anthropic. Fong pointed out that where engineering squads used to have five to ten people collaborating on a project, AI tooling now encourages tiny, solitary teams.

When one engineer can generate the output of four people, companies stop assembling full teams. The banter, the whiteboard sessions, and the shared struggle of debugging an outage together vanish. The engineer sits alone in a room, prompting a model. The output looks high on paper, but the human connection that made hard work sustainable disappears.

Mourning What Was to Build What Is Next

Most tech leaders respond to this anxiety with pure optimism. They tell their teams that AI will remove the boring work so everyone can focus on high-level strategy. Graham calls this out as toxic positivity that alienates the team.

“Change sucks. It also can be awesome, but we don't have to be fluffy bunnies about this. We can also just say this is hard,” Graham said. Her solution starts with letting teams process what they lost. “Let's take a minute and mourn what was, but then let's ask what could be. What could you love that way again? What could you be extraordinary at that way again in this new world if the robots are doing X, Y, and Z?”

Great leadership during technology shifts requires space for that grief. You cannot force people to love steering until you validate why they loved rowing in the first place.

What to Do With This

In your next engineering 1-on-1 this week, do not ask about sprint velocity or AI tool adoption. Ask your report directly: “What part of your day-to-day craft do you miss the most right now?” Listen without offering solutions, validate the loss, and then identify one project where they can work directly with a peer instead of working alone with an AI copilot.