Articles
Guides to routing, reviewing and paying for AI inference, from the team behind Inference Bureau.
Apps
Managed install and token exchange: no more OAuth redirects
Shopify apps on Apps install without an OAuth redirect: Shopify installs them, and the first admin load trades its session token for the store's token.
AppsMerge to main, ship: environments that follow a branch
An app's environment can follow a GitHub branch: a push syncs its files, a merge to main deploys production, and pull requests get a deploy-plan check.
AppsGive your coding agent an MCP server for your app's data
Each Apps environment is an MCP server: your agent reads the app's models and records, runs its actions and reads its logs. How to connect, and what writes.
AppsHow to host a Shopify app in 2026: every piece you need
What a Shopify app needs in production: an embeddable host, auth, verified webhooks, compliance topics, a database, jobs, cron, extensions and alerts.
AppsFailure alerts people don't mute: at most one email per app every 12 hours
When an app's production fails, the team gets an email naming what failed and why. At most one per app every 12 hours, and every failure is in exactly one.
AppsApps is in beta: why we built it, and what comes next
Apps hosts an app's models, actions and pages next to your models, on one bill. Why we built it, what's on the roadmap, and why we want your feedback.
AppsHow we're building hosted apps on Inference Bureau
We're building a place to run the code behind an app, a webhook or a scheduled job, next to your models and on the same bill. Here's why, and what it should fix.
Host apps and functions next to your model traffic. Routing picks each turn's model by your quality floor, and Auditor reviews every pull request against your own rules.