Encore

Hosted platforms like Supabase package managed databases and serverless endpoints with compute add-ons and egress fees that escalate as traffic grows. Encore eliminates hosted platform markups by generating cloud infrastructure directly from Go and TypeScript code into an existing AWS or GCP account. Application code defines databases, queues, and cron jobs directly, letting engineers test locally without maintaining separate Terraform files.

Why Encore instead of Supabase

Supabase provides a preconfigured PostgreSQL backend, pairing authentication, storage, and auto-generated APIs into a managed platform. That model accelerates prototyping, but binds architecture to hosted pricing tiers with surcharges for compute upgrades, egress, and active users. Encore treats backend infrastructure as a compilation target. Services run on AWS Fargate, EKS, Cloud Run, or GKE, backed by cloud resources like Amazon RDS and Cloud Pub/Sub. Infrastructure costs match raw cloud provider rates, letting teams use existing cloud credits without paying a platform intermediary.

The trade-off centers on client-side convenience versus architecture ownership. Supabase excels when frontends query the database directly using Row-Level Security and client SDKs. Encore does not offer browser-facing data access; developers write explicit API handlers and domain logic. When systems scale into asynchronous queues, distributed events, and microservices, Encore avoids the migration hurdles common to BaaS platforms. Teams manage schemas, event topics, and caches in code, avoiding rewrites when requirements outgrow a hosted backend.

What it does well

• Code-defined infrastructure: Developers declare PostgreSQL databases, Pub/Sub topics, Redis caches, and cron jobs directly in Go or TypeScript without hand-written Terraform.
• Local runtime fidelity: Running encore run provisions local PostgreSQL, Redis, and event brokers, ensuring local development matches production semantics.
• Built-in local observability: Every local session includes a dashboard at port 9400 displaying architecture graphs, API request logs, and distributed traces.
• Automated cloud provisioning: Deployments to AWS or GCP translate code graphs into native cloud resources with least-privilege IAM policies.
• Type-safe service calls: Inter-service communication is validated at compile time, eliminating manual API client generation and runtime drift.