Session 2 — Leading a Team

Lunch & Learn series on From Code to Team (Second Edition), Chapter 2. 20 minutes: a story before the first slide (under 4), 12 talk, 4 discussion. Verbatim script; timing marks are the running clock. Slide numbers match the deck.

Room: mixed — leads, aspiring leads, and engineers who never intend to lead.


Before the first slide — the story 0:00

Told with the title slide up and nothing else. Under 4 minutes. Not on any slide. Presenter's voice; swap in your own, keep the shape.

Quick show of hands. Who here has been scuba diving? Keep them up. Who has surfaced somewhere other than where they meant to?

Keep them up. That is more of you than I expected, and we are going to come back to you.

A wreck dive off the coast, a Sunday, 18 meters of green water and a current running. Four of us on the boat. And individually, we were the 4 most experienced divers on it. A hundred dives each, at least. Advanced cards, deep cards, the lot. The skipper knew it. The other groups knew it. We definitely knew it, and we went in off the back of the boat with the easy confidence of people who had done this many times before.

Here is the moment. Fourteen minutes in, somewhere along the port side of the wreck, I look up from a shoal of bream and there are 2 of us. Not 4. Two. I look at my buddy. He shrugs. We look at each other's gauges, look at the compass, and I realize I have no idea which way the boat is — because I had assumed he was navigating, and he, it turned out later, had assumed I was. The other 2 had gone over the top of the wreck 5 minutes earlier, because one of them wanted to see the bow, and the other had followed, assuming the plan had changed. Nobody had said anything. We had all just… been in the water, being good divers.

We surfaced 200 meters from the boat, in the current, 2 pairs, 4 minutes apart, and the skipper had to come and get us — the 4 most experienced divers on the boat, retrieved like students.

The group that went in after us was 3 people who had done maybe 30 dives each. And what I watched them do on the deck, while we were strapping on tanks, was something we had not done: stand in a circle and say it out loud. Max depth 18. Forty minutes or 100 bar, whichever first. You navigate, I watch depth and time, she stays on your left shoulder. If we separate, one minute, then surface. Four sentences. Three divers who had agreed, out loud, what each of them was for. They came up exactly where the boat was.

Afterwards, nobody on our team was angry with anyone. We were all a bit sheepish about the separation, as if it were the current — a thing that had happened to us. And the skipper, coiling a rope, said the sentence I have never forgotten: "You weren't a dive team. You were 4 good divers who happened to be in the same water." Nobody had said what the dive was. Nobody had said whose job navigation was. Each of us had a perfectly good dive, and the dive — the actual dive — belonged to no one.

If you surfaced in the wrong place, put your hand down now. I would bet money nobody said who was navigating on that one either.

That is what today is about. Four brilliant people who report to the same person are not a team. They are 4 good divers in the same water — until somebody says what the dive is, and whose job the compass is.

Slide 1 — Title 3:30

Seattle, early 2022. I inherited the EC2 Outposts team — 19 engineers across 4 functions: hardware supply, field deployment, control-plane software, and ops on-call. They had been under one org chart without a shared leader for close to a year.

Three weeks in, I asked each function lead the same question in 4 separate 1:1s: what is the team's goal this quarter? Hardware supply said: landing the Q1 rack build target. Field deployment: closing the install-defect rate below the reset threshold. Control plane: shipping the auth refactor before the customer commit. And the ops lead, quieter than the others: "I don't know. I've been trying to work that out for a year."

Four answers. None of them wrong on their own patch. None of them the team's. Sitting with those 4 sentences on a page that weekend, I understood what I had actually inherited. Not a team having a bad quarter. A team that had never been asked, out loud, what it was together for. Nobody on that team was broken. The team was.

Slide 2 — A team is not an org chart 5:00