Ethan Evans × Ryan Peterman · Amazon VP: Questions On Corporate Politics But They Get Increasingly Darker
Two hours and twenty minutes in, Peterman asks what "executive presence" is. Evans lists the traits: confidence, decisiveness, specificity. Then he says that in his own executive-presence class he tells students that statistics with numbers are 62% more believable than statistics without them. A beat later:
Now, the truth is I just made that up. I've never looked it up.— Ethan Evans · [02:21:11]
62% sounds specific, he goes on, and I delivered it like it was true, so you probably bought it. A few minutes earlier he had cited a "researcher's" formula for presence (60% gravitas, 30% communication, 10% appearance), and those numbers don't match the source either; see the fact-check below. The moment is the whole episode in miniature: he teaches you to say a little more than the facts strictly support, demonstrates it in front of you, and then exposes himself.
This is Evans's third time on Peterman's show. The first covered the performance machine and PIPs; the second covered "polite fictions" and turf wars between orgs. This time the format is a prepared list of questions, each darker than the last: how to get credit, how to win scope, what to do when there is one promotion slot, what to do when your manager wants you gone, whether he ever burned someone to get ahead. Evans takes them gamely ("Darker? Let's go."), and Peterman keeps coming back with the same follow-up: isn't this another gray area?
Peterman explains up front why he is asking now: models can increasingly do technical work, and if there are fewer tech roles, the same number of people will compete for fewer seats, so politics matters more [00:00:38]. The episode went up at the end of September 2026; earlier that year Amazon had announced about 16,000 more corporate job cuts. The question of the first two episodes, what happens when growth stops, is now the backdrop.
The earlier episodes were about how to say things. Here, for the first time, Evans addresses head-on how far you can stretch them. His example is himself: he always says he was the first engineer on Prime Video, though when he worked on it the product wasn't called Prime yet. Someone once challenged him on it. His answer: "the first engineer on what became Prime Video" is too many words for anyone to remember.
So, I'm ballooning up my claim. It's not false, but it is how you look at it.— Ethan Evans · [00:12:27]
Peterman pushes: there's a spectrum from "truth no one could dispute" to "outright lie"; where should you sit? More is better, Evans says, until it starts to damage your reputation [00:16:33]. His argument is an exam analogy. In college every question gets a score. At work, six months later the manager sits down with a pile of exams, rereads none of them, and thinks, "Ryan? About a 75." Your job is to make sure the manager thinks 97. He is blunt about the cost: a colleague called Alfred quietly waits for fair grading, and next to everyone else's story his 75 becomes an F.
There's a harder version of the technique. To push a manager for a promotion, say you're "in a season of life" with a window, say, planning to start a family, or needing to move home in a few years to care for aging parents. You may not really be planning kids, but as long as you haven't ruled it out, it isn't a lie. And the manager can't push back:
You can't argue with what I feel or what I believe, right?— Ethan Evans · [01:11:59]
He leaves quiet people one route that requires no talking: build something everyone uses. Amazon had an engineer named Matt Round whom Evans never met but everyone knew of, because for more than a decade the interview system was called MRT, the Matt Round Tool [00:19:30]. The thread lands somewhere he doesn't spell out: stretching works because listeners don't check. His 62% is the proof.
The second thread looks at the same thing from the manager's side. There's one good project; Ryan asks for it every day, and Alfred never says a word. Evans says he won't tell himself "I'm shortchanging the nice people." He'll tell himself "I'll make it up to them later," and then never get around to it:
And yes, for all of you who don't like it, I have just said a little bit screw the nice people for the bad people, but I have a job to do.— Ethan Evans · [00:25:42]
So he teaches people to be "thorny": take the bad assignment with a smile, but add, "you only get three crap-work tokens a month, and this is one." The boss never agreed to three, but by not objecting in public he has let the anchor drop [01:37:02]. In the other direction, a manager who wants someone gone has, he says, fifty ways to do it. One: ask Fred to list a hundred useless tickets, pick twenty, hand them to you, and ask you to report on their value. Every step is defensible to HR [01:21:36]. You aren't fired; you "evaporate."
What makes this section interesting is that Evans turns the same knife on himself.
And humans are incredible rationalization machines.— Ethan Evans · [01:42:43]
Asked whether he ever built an empire, he says yes, and that he didn't see it that way at the time:
I think in hindsight I for sure did some empire building but at the time you probably don't tell yourself you're empire building.— Ethan Evans · [01:55:30]
His example is the annual meeting the year the Appstore launched. They asked Bezos for a big batch of people, including "five countries, one PR person per country." Bezos asked two questions, got two answers that fell apart, and said he didn't believe any of the remaining numbers: go back, make a sheet where no bucket is bigger than two or three people, and say exactly what each one will do. Then, with his direct leaders watching, he went through it line by line with a red pen [01:58:00]. Later Evans learned to expand more cleverly. Amazon often opened more requisitions than it had budget for, first come first served, and he always filled his before the gate came down. He would also talk his bosses into lending him headcount that other teams weren't going to fill. He admits that this was taking other people's heads.
Asked whether he ever burned someone to get ahead, he says he's cautious to say no, "because of rationalization." The closest case was his first week: he picked the video project, and another manager who started the next Monday had been promised that very job [02:50:43]. Finally he tells a story that has nothing to do with work. He and a friend gave about $1,000 to a Ukrainian friend so she could bribe her husband, who had served two years, out of the army. Perhaps someone else went to the front in his place, he says, and he is completely okay with that decision [02:56:32]. It is the most extreme statement of his "I deal with the world as it is." He says people rationalize, and in the same breath he calmly gives his own reasons. Whether both can hold at once is a question the show doesn't ask.
Peterman asks: what if everyone played these moves optimally? Evans says that's the prisoner's dilemma: everyone stabs everyone and you end up back at neutral [01:50:38]. His advice isn't to play harder but to change tables and find a game with weaker players. He uses himself as the example: he isn't the best at public speaking, career knowledge, or social media, but almost no one competes with him where the three overlap.
Changing tables is also his answer to the AI era. Suppose you started at entry level at Amazon today and tried to speedrun to VP:
There is no speedrun of that today.— Ethan Evans · [03:02:34]
His reason is growth: during his tenure Amazon went from 14,000 people to 1.4 million, and now it is roughly flat. His route is to spend 18 to 24 months at Amazon, become an AWS expert, and take that label to a high-growth company. Amazon has tens of thousands of L5s; outside, you might be the only person who knows AWS. On AI, he warns against "winning the race off the cliff": the person who only operates the tools is the last one the tools replace. He bets on judgment. Gmail's engineers may be able to write ten times as many features, but nobody needs ten times as many features; someone has to decide which ones. He also argues that demand may grow, as ATMs supposedly didn't eliminate bank tellers. That one is checked below.
The same "human nature doesn't change" drives his view of delayering: it's a cycle, not a new normal. Jensen Huang has sixty direct reports, but they are all extremely senior. Can an ordinary first-line manager go from ten reports to a hundred? He doubts it [02:14:18]. Layoffs get the same narrative treatment: state "my whole org was eliminated" as a quick fact, then pivot straight to what you can bring the next company.
"A researcher found executive presence is 60% gravitas, 30% communication, 10% appearance." [02:18:27]
Supported, but overstated
The source is presumably Sylvia Ann Hewlett's 2014 book Executive Presence. According to Fortune, her research drew partly on a survey of 268 senior executives. Her numbers are 67%, 28% and 5%, and they record what share of respondents named each factor, not a measured breakdown of what presence is made of.
"People thought ATMs would eliminate bank tellers. Didn't happen." [02:10:13]
Supported, but overstated
It was true through the mid-2000s. Bessen found that branches needed fewer tellers, so banks opened more branches; the US had about 608,000 tellers in 2006. Since then mobile banking has shrunk the branch network. The Bureau of Labor Statistics puts 2025 employment at about 339,000 and projects a further 13% decline over the next decade.
"China had civil-service testing by 600 BC, and a formal written exam by 200 BC." [02:12:02]
Supported, but overstated
The second half is roughly right: in 165 BC Emperor Wen of Han was already recruiting officials through examinations. The first half is about 1,200 years early: the standard account dates the imperial examination to the Sui dynasty in AD 605. His conclusion, that management layers are ancient, survives.