A Bestselling Author's Notes: 16 Years in One Vault

Sixteen years of notes from three apps, sorted by whose words they are, so her next book can grow from her own.

Client
A New York Times bestselling author
Industry
Publishing
Scope
Knowledge ManagementData MigrationInformation ArchitectureAI Automation
Stack
ObsidianClaude Cowork
Models
Claude Fable 5.1
Status
Delivered 2026
Notes she sent
18 GB+
Years of notes
16
A bestselling author's second brain in Obsidian's graph view, its notes clustered round their hubs, beside the folder list open on the System folder's dashboards, migration records and templates

The brief#

Silverthread Labs handed a New York Times bestselling author one Obsidian vault that holds sixteen years of her notes, sorted by whose words each file contains, with a Claude prompt that files each new note once she approves.

Her essay collection was a New York Times bestseller, and she also writes a newsletter.

She described the job as small: a pile of notes, most of them in Bear, bound for Obsidian. The night they first spoke, Saif Pasha told her that both apps only read Markdown files, so what mattered was the files and how they were organised.

From the first files she sent, he drew up a plan of the vault, folder by folder. She fixed one detail: her newsletter is not a book. Then she named the real goal. She wanted a home for her own writing, the quick thoughts and longer passages buried among material saved from other writers, where she could ask what she had on a subject and let one idea become a book, a memoir or a newsletter issue. Her next book, already under contract, would be the first test.

The hard part: 18 GB no plain import could copy#

Those first files were not all of it. Notion's export had skipped every page nested inside another, so links across her Notion pages led nowhere. Saif asked for a second export with Create folders for subpages switched on. The full set was more than 18 GB: sixteen years from Bear, Evernote and Notion, as Markdown, PDFs, plain text, HTML and complete Notion pages. Saif reran the whole analysis on it, since that much new material could change the plan.

Nothing could be copied over as it stood. Bear's IDs read like timestamps but put the day first, so a standard import would have shifted notes to the wrong dates without a warning. A Mac zip's __MACOSX folder put a ._ stub beside every real file, doubling any naive count. And her private material sat mixed in with everything else.

[ What each export hid ]

What Saif built#

He built the vault around two facts about a note that never change: whose words it holds, and what stage it has reached. Her own writing sits on top, her books beside a Notebook for ideas with no project yet. Other writers' words go into an archive below that only grows. Topics are tags, not folders, since one note can cover five subjects but can sit in only one folder. Her private material lives in one folder, so a single rule keeps it out of every sync.

She wanted links with a purpose and her files left alone, so each link had to earn a one-sentence reason, and no link went into her notes. The links live in hub notes instead: each subject has one, which lists its notes and the reason for each, and Obsidian's backlinks show the hub from inside each note it names. Beyond a properties block at the top of each note, none of her words were changed, and Saif fingerprinted every original before the work started.

New notes land in an Inbox, and one prompt files them. She opens the vault in Claude Cowork with the triage prompt Saif wrote, and Cowork returns a table of every file, its proposed place and the reason, and holds every move for her approval. When it can't tell her writing from something she saved, it asks, and keeps each answer as a standing rule, so the questions thin out with use.

[ Inside the vault ]

The outcome#

Saif handed her the vault together with How to use your vault, a guide meant for her and for any AI she points at it. She replied that night:

Hurrah! It works!
Esme Wang, New York Times bestselling author

By the handoff call she was running it on her phone and her desktop, in her own colours and fonts, with folders she added herself. After Saif walked her through it, she ran her first prompt on the vault. Within seconds it cleaned up the titles in her Notebook and left the text under each one as it was.

She now has sixteen years of notes from three apps in one vault, sorted by whose words they are, with her own ready to grow into her next book.

What a champ. He made a second brain for me in Obsidian after I threw thousands of files at his head and it came out beautifully.
Esme Wang, New York Times bestselling author

Led by Saif Pasha, with Burhan Pasha. Built by the team at Silverthread Labs. © 2026 Saif Pasha and Silverthread Labs. Ref. SL-STL-SB-2607

Last updated: November 10, 2026

Back to Projects

[ How It Works ]

Free Automation Audit

We find the 20% of your manual work that costs you the most, then show you exactly how to eliminate it.

STEP 1.0
Tell Us What Hurts

Tell Us What Hurts

A 30-minute call. Walk us through your daily operations and we'll spot the bottlenecks you've stopped noticing.

STEP 2.0
We Rank the Wins

We Rank the Wins

We score every opportunity by impact and effort, so you can see where AI saves the most time and money.

STEP 3.0
You Get the Playbook

You Get the Playbook

A prioritized roadmap you can act on. Execute it with us or on your own. Yours to keep either way.