Skip to content

Compliance · Public preview

Random Llama Randomizer

Random selection you can prove.

Guided random draws with a pool snapshot, independently verifiable certificates and PDF records. The v2 web app is available as a preview, with self-hosted and offline desktop options documented.

Web · WindowsVersion 2.0.0-alpha.2Source record · October 2026

Built around a specific job.

Random Llama Randomizer now has a guided v2 workflow: choose the program, configure its rules and start date, paste or import the roster, review the next draw, and retain its results. Each certificate records a seed commitment, HMAC-SHA256 ranking, pool snapshot and hash-chain evidence so another person can recompute and verify it through the app, API or CLI. Annual targets and opening balances use the configured program history. The Python/FastAPI backend serves a React PWA, with self-hosted and offline desktop options. The hosted preview runs on a persistent Postgres database. Sign-out revokes sessions server-side, the v2 member forms and roster imports no longer populate dedicated email or phone fields, and billing and reminder delivery remain configuration steps.

Guided setup, a clear next draw and certificates another person can verify.

  • Seed commitment and pool snapshot make the selection reproducible
  • Public, API and CLI certificate verification
  • Guided program setup, pasted roster and history import
  • Configured annual targets with a retained audit trail
  • Self-hosted and offline desktop deployment options
Ways people use it
  • Set up a program and roster through guided questions
  • Run a random draw and record outcomes
  • Independently verify a draw certificate
  • Retain PDF records and inspect historical draws
  • Import a legacy roster into the v2 workflow

The working scope.

  • Guided program setupChoose intent and rules, set the program start and opening balance, then paste or import people.preview
  • Reproducible random drawHMAC-SHA256 ranking over a committed seed and snapshotted pool.preview
  • Independent verificationVerify certificates through /verify, the API or the Python CLI.preview
  • PDF and history recordsRetain selection results, reasons, PDF evidence and linked historical records.preview
  • Program-year targetsTargets, completed tests and opening balances use the configured program year.preview
  • Hosted, self-hosted and desktop optionsA public web preview plus documented Docker and offline desktop launchers.preview
Feature details and benefits

Guided program setup

Choose intent and rules, set the program start and opening balance, then paste or import people.

  • Begin with the next concrete action

Reproducible random draw

HMAC-SHA256 ranking over a committed seed and snapshotted pool.

  • Verify how the selection was made

Independent verification

Verify certificates through /verify, the API or the Python CLI.

  • Recompute the result
  • Reject altered evidence

PDF and history records

Retain selection results, reasons, PDF evidence and linked historical records.

  • Review what happened after the draw

Program-year targets

Targets, completed tests and opening balances use the configured program year.

  • See the next required action

Hosted, self-hosted and desktop options

Hosted customer use still requires persistent database configuration; desktop and self-hosted options use their own configured data storage.

Project details.

Random Llama Randomizer specifications
PlatformWeb · Windows
AvailabilityPublic preview
Version2.0.0-alpha.2
Source recordOctober 2026Synchronization date, not a capture or launch date
Who it is forSafety managers running regulated or non-regulated testing pools · Organizations that need reproducible selection evidence · Teams evaluating self-hosted or offline draw workflows
Tools behind the buildPython · FastAPI · React · TypeScript · Vite · SQLAlchemy · SQLite · Postgres · PyInstaller

Public update log.

Recorded update · 2026-09-26 · 2.0.0-alpha.2

A guided v2 preview with verifiable draw certificates

Randomizer's v2 web workflow guides program setup, roster import, draws and evidence verification. The public preview runs on a persistent Postgres database as of October 9.

  • Guided questions lead to the first draw
  • Program start and opening balances support full-year targets
  • Pasted and CSV roster imports preserve recorded identities and statuses
  • Sign-out revokes sessions server-side; old tokens stop working
  • The v2 member forms and roster imports no longer populate dedicated email or phone fields
  • Verification recomputes the committed seed, ranking and pool snapshot
  • PDF evidence and legacy roster import are available
  • Hosted billing and reminder delivery remain configuration steps
Random Llama Randomizer updates
October 2026Session revocation and a leaner member recordSign-out now invalidates the session token server-side, so a copied cookie stops working the moment you sign out. The v2 member forms and roster imports no longer populate dedicated email or phone fields. The hosted preview runs on a persistent Postgres database.
September 2026A guided v2 preview with verifiable draw certificatesRandomizer's v2 web workflow guides program setup, roster import, draws and evidence verification. Deployment accepts a connected Postgres database.

A useful idea of your own?

Start with the business task. We will work out what needs building and what belongs in the scope.

Discuss your project

Thursdays

Get the digest on Thursdays

Each Thursday: the shop notes and AI daily issues from the last seven days, and one product. Unsubscribe from any issue.