[Ethan's Apartment — July 2015, Late Night, 11:30 PM]
The file was on his laptop. A text document, created January 15, 2014 — the second day of his second life, typed on the dead man's keyboard while Folgers brewed in the dead man's kitchen and the reality of transmigration settled into his consciousness like sediment in still water.
SHOW_NOTES.txt
The document was forty-three pages. Everything he'd remembered about Silicon Valley — the HBO show, the plot, the characters, the events that defined the six seasons he'd half-watched in his previous life. Richard Hendricks' compression algorithm. Pied Piper's legal battles with Hooli. Gavin Belson's megalomaniac leadership. Erlich Bachman's incubator. Jian-Yang's schemes. The rise and fall and rise of a startup that had been the show's through-line across sixty episodes of comedy and catastrophe.
He'd written it in a rush, terrified of forgetting, aware that the details were already fading — the particular unreliability of memories formed through casual viewing rather than study. Show knowledge was entertainment knowledge, the kind of information that stuck in fragments: iconic scenes, memorable quotes, plot twists that had surprised him during a Tuesday-night binge. The granular details — episode-by-episode events, specific dates, secondary character arcs — had been sketchy from the start and had degraded with every month in the timeline.
Ethan scrolled through the document. The entries were dated, annotated with confidence levels he'd assigned eighteen months ago:
Richard wins TechCrunch Disrupt — HIGH CONFIDENCE (major plot point, Season 1) Pied Piper vs. Hooli arbitration — MEDIUM CONFIDENCE (happened in Season 1 or 2, details fuzzy) Peter Gregory dies — HIGH CONFIDENCE (Season 2 opening, specific and memorable) Gavin launches HooliChat or similar — MEDIUM CONFIDENCE (some kind of failed product, details uncertain) Erlich gets involved with various schemes — LOW CONFIDENCE (specific schemes unclear) Jian-Yang does something with a food identification app — MEDIUM CONFIDENCE (hot dog app? SeeFood?)
The confidence ratings were themselves artifacts of a time when the show's plot had felt like a map — imperfect but useful, a guide to the terrain he was navigating. He'd used the file in the first months: avoiding the Basecamp pitch meeting on the day when, according to his notes, a particular VC firm had been in a bad mood due to a portfolio company's failure. Timing the TechCrunch Disrupt attendance to coincide with Richard's presentation. Approaching Monica before she was fully committed to Pied Piper.
Those early interventions had worked because the timeline was still close to canon. His presence had been a small perturbation — a new character added to the show's cast, operating at the margins, leveraging meta-knowledge for positioning without altering the fundamental plot.
But that was eighteen months ago. The timeline had since diverged to a degree that made the show notes obsolete.
He opened a browser. Pulled up the tech news he'd been tracking — the aggregated landscape of the industry he was reshaping. The headlines told the story of divergence:
"AI Investment Surges 300% as Silicon Valley Bets on Machine Intelligence" — Bloomberg "Google Brain Accelerates Transformer Research Following Gardner Analytics' Groundbreaking Paper" — Wired "The Gardner-Pied Piper Partnership: How Two Startups Are Outpacing Big Tech" — The Verge "Hooli Announces $500M AI Research Initiative Led by VP Vincent Mora" — TechCrunch "DeepMind Shifts Timeline for Natural Language Understanding Program" — Nature
Every headline represented a butterfly effect. His existence in this timeline had accelerated AI research by three to five years. The Transformer paper — published in 2014 instead of 2017 — had given the academic community a foundation that was already producing derivatives and extensions. Google Brain was building on his architecture. DeepMind had adjusted its research priorities. Hooli had committed half a billion dollars to an initiative that wouldn't have existed without the competitive pressure his company created.
The Silicon Valley he remembered from the show — the one where Pied Piper struggled against Hooli while the broader tech landscape focused on mobile apps and social media — was gone. Replaced by a Silicon Valley where AI was the defining investment thesis, where his company was a category leader, where the competitive dynamics had shifted from the show's comedy of errors to a genuine technological arms race.
The show notes were useless.
He scrolled to the bottom of the document. The last entries, written in February 2014, covered events he'd predicted for the second and third seasons:
Pied Piper grows, faces more legal challenges — MEDIUM CONFIDENCE Gavin eventually leaves Hooli or gets pushed out — LOW CONFIDENCE Someone (Jian-Yang?) does identity fraud or corporate espionage — LOW CONFIDENCE The company eventually goes public or gets acquired — LOW CONFIDENCE (don't remember ending clearly)
Four predictions, all rated LOW or MEDIUM confidence, none of them applicable to a timeline where Gardner Analytics existed, where the Pied Piper-Gardner partnership had reshaped the competitive landscape, where Vincent Mora was running a $500 million AI initiative that the show had never imagined.
Ethan selected all forty-three pages. The cursor blinked in the text field, the universal prompt for a decision: keep or delete.
He deleted the file. Emptied the trash. The forty-three pages of show knowledge — the map that had guided his first months, the prophecy that had degraded into irrelevance — disappeared from the laptop's filesystem with the finality of a file that would not be recovered.
The apartment was quiet. The same apartment — the dead man's apartment, his apartment now, the space he'd lived in for eighteen months without changing the furniture or painting the walls or doing anything to make it feel less like a borrowed room and more like a home. Monica had been suggesting he move. He'd been resisting because the apartment was cheap and convenient and, underneath those practical justifications, because leaving it felt like severing the last connection to the person he'd been when he arrived.
But the person he'd been when he arrived was a transmigrant following a script. The person he was now was a CEO improvising.
The meta-knowledge that remained was technical: the architectural blueprints, the scaling laws, the progression from Transformer to GPT to GPT-2 to GPT-3, the fundamental mathematics of attention and generation that existed independently of any show's plot. Those would stay. Those were supernatural, anchored to the ability rather than memory, refreshed through implementation and training rather than decaying through disuse.
Everything else — character knowledge, plot predictions, timeline expectations — was gone. Not just deleted from the laptop. Gone from relevance. The show he'd watched in a different life described a universe that no longer existed, populated by people who'd been changed by his presence in ways the show could never have anticipated.
Richard Hendricks was a partner now, not just a protagonist. Monica Hall was a CFO and a lover, not just an investor. Gilfoyle was an ally, not just a side character. Gavin Belson was a personal enemy with a $500 million budget. Vincent Mora was an adversary who didn't exist in any episode Ethan had ever watched.
The future was genuinely unknown. Not approximately unknown, not fuzzy-around-the-edges unknown, but the deep, structural unknown of a timeline that had diverged past the point of useful prediction. Ethan Gardner was flying blind. The architectural awareness would guide the technology. Everything else — business strategy, competitive dynamics, personal relationships, the thousand daily decisions that determined whether a company survived — would require the same improvisation that every other CEO in Silicon Valley practiced without supernatural assistance.
The thought was terrifying. The thought was also, in a way he hadn't expected, liberating. The meta-knowledge had been a crutch — a crutch that degraded with use, that produced diminishing returns, that created a false confidence in predictions that were increasingly wrong. Without it, the decisions were his. The mistakes would be his. The successes would be genuinely earned rather than extracted from a show he'd half-remembered.
He closed the laptop. The apartment was dark except for the streetlight filtering through the window — the same light from the first morning, the same angle, the same quality that San Francisco's particular geography produced at the intersection of urban lighting and coastal fog.
The dead man's water stain on the ceiling was still there. He'd stopped noticing it months ago, the way people stop noticing the permanent features of their environment. But tonight it was visible — a Rorschach in the plaster, the same shape he'd stared at on his first morning while the enormity of transmigration pressed down on him like atmospheric weight.
He looked at it. The stain looked back. Neither of them had changed. Everything else had.
His phone buzzed. Monica, from her apartment across the city.
Monday went well. Laurie approved the transition. I start at Gardner in thirty days. The portfolio review wasn't about cuts — it was about doubling down on winners. We're a winner.
Relief mixed with something warmer. Monica was coming. The company was safe at Raviga. The future — the genuinely unknown, genuinely unscripted future — was beginning.
He typed back: Welcome to the team. Your first assignment: help me figure out how to raise $50M for a training run.
Three dots. Monica typing.
$50M? What are you building?
The next generation. I'll show you the whiteboard Monday.
He set the phone down. The apartment held its silence. Somewhere in the building, pipes hummed — the infrastructure of a city that didn't know it was being reshaped by a man who'd stopped following a script and started writing one.
Author's Note / Promotion: Your Reviews and Power Stones are the best way to show support. They help me know what you're enjoying and bring in new readers! You don't have to. Get instant access to more content by supporting me on Patreon. I have three options so you can pick how far ahead you want to be: 🪙 Silver Tier ($6): Read 10 chapters ahead of the public site. 👑 Gold Tier ($9): Get 15-20 chapters ahead of the public site. 💎 Platinum Tier ($15): The ultimate experience. Get new chapters the second I finish them . No waiting for weekly drops, just pure, instant access. Your support helps me write more . 👉 Find it all at patreon.com/fanficwriter1