One platform, one API, one operations model. The secure backend runs underneath — you build the product on top.
You design the model and the workflows. Jetstack runs them — with hosting, queues, files, AI, and audit-grade observability already wired in.
The building blocks teams use day-to-day to ship business apps fast and keep them predictable.
Define object prototypes and properties with explicit data types and constraints. The core handles many lower-level invariants implicitly so your team focuses on model quality, not boilerplate.
Structured views for analysis and operational workflows. Inspect details in context windows, group, filter, search across data with full-text, and export to native XLSX without leaving the page.
Compose pages and dashboards from blocks or custom HTML. Embed any data view, anywhere — no need to settle for table-only screens.
A what-you-see-is-what-you-get workflow builder. Branch on data, drop in JavaScript, call AI, and iterate over large data streams — or keep it to a single-step notification. Triggers fire from object events, schedules, webhooks, or manual actions.
Connect in every direction. A built-in MCP server lets the team talk to your data from Claude, Cursor, or any client; custom connectors and external MCP servers bring outside data and apps in. Plus a REST API that reflects model changes on the fly, webhooks for outbound events, and public pages for shareable read paths.
Unlimited role definitions, implicit-deny by default, granular operation rights down to single records.
White label is a first-class capability. Customize logos, colors, layout, welcome screens, and email templates from settings.
Most apps never need this. When yours does, you don't migrate off the platform — you go deeper into it. Host your own frontend and custom backend logic right here, with direct access to the same internal APIs the platform runs on.
Host a React or Vite app on the platform. It inherits identity, authentication, and data access — so the UI stays focused on the experience.
A purpose-built backend-for-frontend layer. Sandboxed server code with a stable internal interface to platform data and services.
Your code talks to the same internal APIs the platform runs on — records, files, search, permissions, queues — without round-trips through a public gateway.
A built-in code editor. No separate toolchain to wire up — change code where it runs.
Code is compiled and packaged into an optimized production build. No build servers for you to maintain.
Every code deploy is tracked and can be stepped back if a release misbehaves — all inside a controlled boundary that preserves the platform's security guarantees.
AI runs through the platform, not bolted on the side — from an in-app assistant to autonomous agents that act through the same governed tools and permissions as everyone else.
An in-app assistant that understands your data model and helps people navigate, build, and configure — without leaving the app.
Expose your data to Claude, Cursor, or any agent through a built-in MCP server — and reach outward to external MCP servers your workflows can call.
Autonomous agents that work a task across multiple steps, using the platform's own capabilities as their action surface — governed by the same permissions.
A visual builder for workflows: branch on data, drop in JavaScript, call AI, and iterate over large streams — or stay single-step simple.
Connect to external data and apps with custom connectors — bring third-party services into your model and your automations.
Let AI accelerate the slowest part of implementation — shaping the model and drafting automations.
The infrastructure layer most teams build, run, and pay for separately — already inside Jetstack.
Production-grade hosting, scaling, and delivery — managed.
Workloads run on AWS. Inherits AWS-grade reliability and certifications.
Defined tiers per app. Match cost to actual needs and switch on demand.
Ship app changes through the platform — no build pipelines to maintain.
Edge delivery, automatic certificates, secure-by-default endpoints.
A stable, predictable API surface that reflects your model.
Reflects model changes on the fly. Predictable filtering and pagination.
Fast, indexed search across object data and properties.
Per-key, per-route limits. Transparent quotas and headers.
Outbound events with signing, retries, and delivery logs.
Run work later. Tell devices when it happens.
Durable queues for retryable async work with backoff and dead-letter visibility.
Schedule actions for an exact time. One-shot or recurring.
Subscribe terminal devices to a server-sent event stream out of the box.
Built-in push channel — including notifications to mobile devices.
A real file repository with secure delivery, not just blob storage.
First-class files attached to objects with metadata, versions, and auth.
Time-bounded, tamper-proof links for controlled access to private files.
Resize, convert, sign, and stream files without external services.
Publish read-only views or content on shareable URLs.
See what happened. Prove who did it.
Configurable audit log built for compliance and transparency obligations.
Health, throughput, and error visibility per app.
Stream platform and app logs without leaving the console.
Implicit-deny model, field- and record-level scoping.
Talk to your data via MCP from the environment you already use, plus a code-or-tools automation builder and AI agents.
Talk to your data from Claude, Cursor, or any MCP environment — read, query, and trigger actions through the tools you already use.
Compose workflows from a rich tool library — or drop into code.
Build and run agents with platform tools as their action surface.
Object events, schedules, webhooks, manual actions — with branching and retries.
In-app, email, and push notifications. Custom SMTP relay supported.
Drop widgets onto dashboards and modules. Custom programming where you need it.
Flat entry, then a usage-based curve tied to data — no surprise dimensions.
The core is generic and fits broad system types — including large, complex apps.
Move from technical modules to a delivery plan, ownership model, and production timeline.
Open implementation tracks, docs, and API references and start shipping.
Share your use case. We'll show what fits and what to leave alone.
Tell us about your project. We will reply shortly.
