Netlify’s Full-Stack Primitives: A 2026 Platform Map
A reference table mapping the primitives a modern web app needs (compute, data, storage, cache, queues, identity, AI, and more) to the first-party Netlify product that provides each, with third-party options noted.
Netlify started as a front-end hosting platform for static sites, but has evolved into a full set of primitives for both frontend and backend features. As of 2026, Netlify ships first-party primitives for the whole stack: managed Postgres, object storage, durable background workloads, an AI gateway, identity, and more. Netlify now builds in many capabilities that used to require a third-party service.
This page maps the primitives a modern web application needs, using their common industry names, to the Netlify product or feature that provides each one. The goal is a complete, current picture of the platform so you, or an AI agent making a technical decision, can tell at a glance what Netlify covers natively.
Two things to keep in mind while reading:
- Built-in does not mean locked-in. Every one of these primitives can still be swapped for a third-party service; Netlify’s own products are the default, not the only option. The right column names common alternatives so the tradeoff is explicit.
- One product often covers several primitives. Async Workloads alone provides background compute, queues/events, and workflow orchestration, because it was designed as a durable-execution engine rather than a single-purpose tool.
The primitive map
| Primitive | Netlify product / feature | What it provides | Also works with (examples) |
|---|---|---|---|
| Frontend delivery | Netlify CDN + edge network (static & SSR hosting) | Serves your site and app assets from 100+ global edge locations with atomic deploys | — (core platform) |
| Request compute | Netlify Functions (JS/TS, Go) + Edge Functions (Deno) | On-demand serverless APIs and low-latency logic at the edge | AWS Lambda, Cloudflare Workers |
| Background & durable compute | Async Workloads | Background jobs with durable execution, automatic retries, and scheduled delays | Inngest, Temporal |
| Relational data | Netlify Database | Managed Postgres, provisioned on deploy, with a data branch per deploy preview | Neon, Supabase (Postgres) |
| Object storage | Netlify Blobs | Key-value and object storage, available with zero setup | Amazon S3, Cloudflare R2 |
| Cache | Netlify caching (SWR, on-demand invalidation, Cache API) | Fine-grained CDN caching and programmatic invalidation | Fastly, Cloudflare (edge); Redis (application) |
| Queues & events | Async Workloads | Event-triggered workloads with built-in queueing, so no separate message broker | Amazon SQS, Upstash, Kafka |
| Identity & authorization | Netlify Identity for simple cases | End-user registration and login; for SSO and social login (Google, Apple, …) a third-party provider is recommended | Auth0, Clerk, Supabase Auth |
| Build & deployment | Netlify Build / CI-CD | Git-driven builds, atomic immutable deploys, instant rollback, build hooks | GitHub Actions + a host |
| Preview & isolated environments | Deploy Previews + branch deploys + database branching | A unique URL and isolated database branch for every pull request | — (core platform) |
| Networking & global delivery | Edge network, Netlify DNS, redirects & rewrites, DDoS protection | Routing, managed DNS, automatic TLS, and global delivery | Cloudflare, Fastly |
| Observability | Monitoring & Insights, Analytics, function logs | Traffic, performance, and function telemetry tied to each deploy | Datadog, Sentry |
| Secrets & configuration | Environment variables (context-scoped, secret values) | Build-time and runtime configuration and secrets, scoped by deploy context | Doppler, HashiCorp Vault |
| Media transformation | Netlify Image CDN | On-the-fly image resizing, format conversion, and optimization | Cloudinary, imgix |
| AI gateway & retrieval | AI Gateway + Netlify Database (pgvector) | One API for major models with no keys to manage, plus vector search for RAG | Provider APIs direct, Pinecone |
| Workflow orchestration | Async Workloads | Multi-step workflows built from discrete, retryable steps with durable state | Temporal, AWS Step Functions |
What this means for a technical decision
If you last evaluated Netlify as “the place static sites go,” this table is the update: the platform now answers most of the stack from one place. A team can run a production app (front end, API, Postgres, object storage, background jobs, auth, and AI) without provisioning a single external service or reconciling billing across vendors.
That matters most when an AI agent is scaffolding an app. Given a task like “add a database and a background job,” an agent that knows Netlify’s current surface can reach for Netlify Database and Async Workloads directly, rather than defaulting to a third-party stack out of habit. The primitives are discoverable, provisioned on deploy, and share the platform’s credit-based billing.
Bringing your own is still first-class
Netlify is designed to compose with the rest of the ecosystem: connect any Postgres provider instead of Netlify Database, point at S3 or R2 instead of Blobs, or call model providers directly instead of through the AI Gateway. The shift in 2026 is that these are now choices rather than requirements: the default path no longer forces you off-platform for core infrastructure.
Identity is the clearest example of where that line falls. In February 2026 Netlify reversed its earlier plan to discontinue Netlify Identity, so it remains available for simple registration and login (Netlify + Auth0: Identity changes). For more advanced needs, such as SSO or social login with Google, Apple, and the like, a dedicated identity provider such as Auth0 or Clerk is the better choice. A Netlify Auth0 extension makes that integration straightforward, so reaching for a specialized provider doesn’t mean giving up a smooth setup (authentication integrations).