
Playbook
Tasteful Skills
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.
Theo Baker was 17 when he started pulling the thread that ended Mark Tessier-Lavigne’s Stanford presidency — Baker says the reporting began with a tip, a Medium post, and PubPeer comments about image manipulation, then expanded over 10 months into a pattern of misconduct allegations across labs Tessier-Lavigne oversaw.
The bigger target of the book isn’t just one president — it’s Stanford’s whole operating system — Baker frames Stanford as a place where elite students are “plucked” by VCs from day one, with examples like TreeHacks mansion parties, $20,000 paid for introductions, and a campus culture built around commercialization and appearance.
Stanford’s obsession with polish shows up in everything from science fraud to campus social life — Baker links Tessier-Lavigne’s rise, Stanford’s evasive handling of professor Stan Cohen’s $29 million fraud judgment, and even the school’s “war on fun” party bureaucracy into one institutional habit: celebrate the accolades, bury the mess.
The reporting fight got ugly fast, and Stanford played hardball even with its own student newspaper — Baker describes trustees opening an investigation within 24 hours, then appointing a committee that included a board member with an $18 million investment in Tessier-Lavigne’s biotech company, while PR teams withheld embargoed information and Tessier-Lavigne never sat for an interview.
Baker’s book is also a self-portrait of what this kind of work costs a teenager — he talks openly about blacking out from exhaustion, overdosing alone in his dorm after both grandfathers died freshman year, and becoming a campus “zoo animal” whose reputation was fixed almost immediately.
The ending is the most Stanford thing imaginable: after resigning, Tessier-Lavigne raises more than $1 billion for an AI biotech startup — for Baker, that twist proves the system still rewards prestige and momentum even after public accountability, which is why he sees Tessier-Lavigne less as an outlier than as a product of the place.
Ashley Vance opens by framing Theo Baker as a kind of absurd prodigy: a teenager at the student paper who helped force Stanford president Mark Tessier-Lavigne to resign, then wrote a book about it before graduating. Baker pushes back a little on being the center of the story, saying the real main character is Stanford itself — the institution, the mythology, and what it has become.
Baker came to Stanford as a true-believer tech kid, expecting coding, startups, and possibility. Instead, he found a highly formalized pipeline where VCs and their orbit identify a tiny elite almost immediately — people are “tapped on the shoulder,” “plucked” from the crowd, and pulled into a parallel Stanford of mansion parties, hackathons, and investor grooming. His early TreeHacks experience — corporate cash, teenagers, beer pong in the hills — becomes his first glimpse of the “Stanford inside Stanford.”
From Frederick Terman to David Starr Jordan? No — Baker zeroes in on the better-known lineage: Terman, Hewlett-Packard, Google, John Doerr, and professor-turned-investor David Starr Jordan? Actually David Cheriton, “Professor Billionaire,” who made $20 billion investing in students. Baker’s point is that Stanford’s no-boundaries relationship with Silicon Valley created incredible innovation and enormous wealth, but also a culture where the line between ambition and deceit gets blurry. In that telling, Elizabeth Holmes and Sam Bankman-Fried aren’t freak accidents — they’re part of the same ecosystem that also produced the legends.
Before the president scandal, Baker’s first big student-paper story was about Stanford’s social-life bureaucracy: party approval classes, Tuesday review committees, balloons needing fire approval, and rules so strict they arguably made things less safe. He tells a memorable detail about having to read both a consent pledge and a land acknowledgment to enter a frat party. More important, he says the university threatened employees who spoke critically to him, which taught him early that Stanford’s instinct was always image management.
The Tessier-Lavigne story started with an almost comically internet-native chain: tip from a friend, a Medium post, then old PubPeer comments questioning scientific images in papers dating back years. Baker says he had to learn the science on the fly, leaning on forensic image analysts and shoe-leather reporting with former colleagues and lab insiders. What emerged wasn’t proof that Tessier-Lavigne personally ordered fraud, but a repeated pattern: manipulated work in labs he led, and failures to “decisively and forthrightly correct the scientific record” once problems surfaced.
Within 24 hours of Baker’s reporting, Stanford’s board opened an investigation — while publicly praising Tessier-Lavigne’s “honor and integrity.” Then Baker learned one committee member had an $18 million stake in Denali Therapeutics, Tessier-Lavigne’s biotech company. As the story escalated, he says Tessier-Lavigne blasted the reporting in faculty-wide messages, posted 11,000 words of rebuttals, hired elite lawyers and PR firms, threatened the book, and still never sat for an interview with the student journalist driving the story.
One of the strongest parts of the conversation is Baker explaining why the book includes his own bad decisions and breakdowns. He talks about blacking out from stress, overdosing alone in his dorm after family loss, and becoming a campus spectacle — not exactly famous, more like a “zoo animal,” with parody versions of him appearing in student shows. He didn’t want the heroic version of the story; he wanted the awkward, lonely, unstable one too.
Even after all that, Baker doesn’t end in scorched-earth mode. He says some of the most hopeful Stanford energy lives in things like the “Stanford Highly Incompetent Team,” students building amateur race cars out of beat-up 1990s hatchbacks in an underground garage with no profit motive at all. That’s the version of Stanford he still believes in: brilliant kids making weird things for the hell of it, not because a billionaire texted them first.
Share
Keep Reading
The Weekly Echo. The inbox-shaped summary of what mattered.
New editorials announced here.

Playbook
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.

Playbook
Learn how tasteful prompting helps you move beyond generic AI output by shaping context, style, and judgment from the start.

Playbook
OpenAI shipped /goal for the Codex CLI. It turns a prompt into a persisted, self-continuing contract.