Vajra Regent Library & Archives

Vajra Regent Ösel Tendzin Library and Archives is a non-profit that preserves the life and teachings of the Vajra Regent, who taught from 1943 to 1990, for Buddhist communities around the world. Some of its audience studied with him directly, decades ago. Others are discovering his teachings for the first time.

This archive is not a side project. It is the main way his work survives and reaches people who never met him.

For almost ten years, the Library and Archives ran their digital platform, Timeline, to keep that legacy accessible. The platform stopped keeping up with the archive's growth — and with its mission. Here's what had piled up:

• An outdated, insecure stack. Every update got harder and riskier than it needed to be.
• No backup for archive media. One server failure could have erased ten years of work.
• A decade-old interface and limited search. New visitors struggled to find their way in, and the archive's reach stayed limited to people who already knew where to look.
• No self-service tools for admins. Routine content and design changes always needed a developer.
• Much of the archive existed only as audio. Hundreds of hours of teachings had no transcripts, so none of it could be searched. Anyone unwilling to sit through hours of recordings to find one passage simply couldn't.

We rebuilt Timeline from the ground up on Rails 8, Hotwire (Turbo and Stimulus), and Tailwind CSS — one release, with every accumulated problem fixed at once instead of spread across a string of smaller updates.

Our designers redesigned the public archive and the admin panel both. The decade-old interface is gone. What replaced it works for visitors and for the client's own team.

The new timeline organizes videos, audio, transcripts, poems, photographs, and calligraphies by the periods of the Vajra Regent's life, so a first-time visitor can follow his path through the years — not just search for what they already know exists.

We added phrase search, month delineators in the table of contents, and page-level content blocks, so both newcomers and long-time students can get to a specific teaching in seconds.

The rebuild changed how the archive is run, not just how it looks:

• Media files moved off the production server into secure cloud storage with backups — a server failure can no longer erase the archive
• Admins manage the landing page, banner, and content themselves through a WYSIWYG editor, without waiting on a developer
• Access-level requests are automated. What used to be manual back-and-forth for admins now runs on its own

Timeline is a digital archive of teachings, and making every recording as searchable as a written page is part of what that means. 937 historical recordings had no transcript. We ran them through OpenAI's Whisper API, with safeguards built in for large files and for cost, and turned what would have been months of manual transcription into weeks of automated processing.

• All 937 recordings are transcribed now. They sit in the same search index as everything else, so the archive is fuller than it was before — not just in theory
• Names and terms specific to the archive, like Ösel Tendzin and Vajradhatu, come through correctly. This material deserves precision
• Costs stay predictable. The system doesn't pay twice for a recording it already processed, and a corrupted file won't turn into a runaway bill

Nearly a decade of use had built up a large archive and a 7-level permission system controlling who sees what. We planned one controlled cutover, with a full rollback plan ready in case something broke.

We carried over the entire archive and the full access logic, one to one, then switched every user to the new version. Zero downtime. Not a single lost record. Once the new platform passed validation, we decommissioned the old one.

Work with the Library and Archives continues:

• Transcription and search for PDF materials. A large part of the archive sits in PDF documents, still outside search. We're ready to extend the same approach we used for audio
• Podcast production from the audio archive. We're scoping the archive's next chapter: turning select audio teachings into a real podcast, with clean metadata on every file and RSS feeds listeners can subscribe to, seminar by seminar

Ten years ago, Timeline was the only way to keep the Vajra Regent's teachings alive online. Over time, it turned into a liability nobody could see: media with no backup, an interface holding the whole archive back, hundreds of hours of teachings that existed nowhere in writing. None of that shows until the day it fails.

Today the archive is preserved. It's searchable in full, including teachings that used to live only in audio. And the Library and Archives can shape how their own work looks and reads, without waiting on anyone else to do it for them.