VIN Doc Platform Release Notes: Mid-2026
This quarter we focused on the things heavy integrators ask for most: more throughput, better introspection, and fewer surprises. Here is what shipped, why it matters, and what is on deck. Everything below is live in production unless noted otherwise, and as always, none of it requires you to rewrite a working integration.
Higher bulk throughput
The bulk endpoint now processes larger jobs with shorter end-to-end times, and completion webhooks fire sooner after the last VIN resolves. If you backfill large inventories, you should see noticeably faster turnaround without changing a line of code, because the gains are server-side.
- Larger maximum job sizes on eligible plans
- Faster completion webhook delivery
- Per-VIN status now includes a confidence field
The new per-VIN confidence field is the one change worth reading for: it lets you route borderline matches to review automatically instead of treating every returned VIN as equally trustworthy.
Richer SDK ergonomics
The official SDKs gained built-in idempotency-key handling and automatic, jittered retries on transient errors. New users inherit safe defaults; existing users can opt in. Pagination helpers now stream results so you can process a large response without buffering it all in memory, which matters when a single page of results is larger than you want to hold at once.
These ergonomics are not cosmetic. Built-in idempotency and jittered retries are exactly the patterns we have written about repeatedly as integration best practice; having them in the SDK means the safe path is now also the default path, and a team can inherit correct behavior without reading a single guide first.
Dashboard introspection
The dashboard now shows per-key usage, latency percentiles, and recent errors in one view, so you can debug an integration without leaving the browser. Webhook deliveries are visible with their full retry history, which makes diagnosing a flaky endpoint far quicker: you can see exactly which delivery failed, how many times we retried, and what your endpoint returned.
- Per-key usage and latency in the dashboard
- Full webhook delivery and retry history
- Inline error inspection for recent failures
The per-key view also makes multi-tenant rollouts easier to reason about. If you have scoped credentials per location, per customer, or per service, you can now see at a glance which keys are busy, which are idle, and which are throwing errors, without exporting logs or building your own dashboard first. That visibility shortens the loop between a problem appearing and someone noticing it.
Why these changes are safe to adopt
Every change this quarter is additive. The confidence field is new, not a replacement; the SDK retries are opt-in for existing users; the dashboard views read data you already produce. Nothing here changes the schema contract you build against, which is the whole point: you adopt improvements on your timeline, not ours.
This is the same additive-change promise that lets a contract test stay green across releases. If you gate your deploys against the schema, none of this quarter's work should turn that gate red; the new fields simply appear alongside the ones you already read. We hold ourselves to that contract precisely so that platform progress never becomes your migration project.
Pricing is unchanged
The offer stays exactly as it was: a two-day trial for €3.99, then €49.99/month, auto-renewing and cancelable anytime, in euros. New throughput and ergonomics come at no change to what you pay, so there is nothing to reprice and nothing to renegotiate to benefit from this quarter's work.
Coming next
We are working on a streaming changes feed for very high-volume fleet sync and finer-grained scopes for enterprise key management. As always, schema changes will be additive, and we will announce breaking changes well in advance. Tell us what would help your integration most: the roadmap is shaped by what heavy integrators actually ask for.
