Supabase engineering blog

Supabase introduces Multigres for PostgreSQL connection pooling and automated failover, OrioleDB for undo-log storage without table bloat or routine VACUUM, and dbarena for reproducible provider benchmarks. The post explains their architecture, scaling trade-offs, availability model, transaction-ID handling, and reported performance results.

Supabase’s 2026 Select recap introduces code-first schema and configuration workflows, native local development, agent-integrated health checks, MCP support, and database observability tools. It also covers new scaling options, including Multigres high availability, OrioleDB, and reproducible database benchmarks.

Supabase introduces tools for coding agents to observe projects, investigate production issues, test fixes, and operate within scoped permissions. The update adds SQL-accessible logs, health checks, connection diagnostics, notebooks, safer MCP controls, and near-real-time Postgres pipelines to analytical destinations.

Supabase is now available as a prebuilt connector in Google Cloud Gemini Enterprise, enabling natural-language queries and actions against live Supabase projects. The post explains edition-specific setup, organization-level connections, real-time data retrieval, and read-only versus destructive tool annotations.

Supabase introduces generally available enterprise-managed authentication for its MCP server, built with Anthropic and Okta. The integration centralizes authorization through Okta while preserving each employee’s existing Supabase roles, project access, onboarding, and offboarding controls.

Supabase’s supabase-js client can now propagate W3C Trace Context into Supabase API Gateway and Edge Function logs, linking client and server activity with a shared trace_id. The post explains setup with OpenTelemetry, sampling options, bundle behavior, log drains, and current limitations.

Supabase is now available as a Perplexity Computer connector, allowing users to query and update Postgres data, inspect user activity, invoke Edge Functions, and schedule recurring workflows. The integration connects Supabase with tools such as Stripe, GitHub, and Slack for multi-step operational tasks.

Supabase announces updates to Postgres Changes, including AND filters, new filter operators, and column selection. The post body was unavailable, so the implementation details and practical trade-offs could not be assessed.

OrioleDB Public Alpha announces the availability of an alpha release of a PostgreSQL-compatible storage engine. The post body was unavailable, so no further technical details or takeaways could be verified.

The post appears to cover executing dynamic JavaScript code within Supabase Edge Functions, likely addressing a serverless edge-execution pattern. The body was unavailable, so this summary is based on the title alone.

The post announces what is new in pgvector v0.7.0; the body is unavailable, so its specific technical changes and takeaways cannot be verified.

This post discusses techniques for minimizing PostgreSQL bloat, a database-maintenance issue that can affect storage use and query performance. The body was unavailable, so this summary is based on the title and URL alone.

The post appears to examine Supabase's SLA Buddy, an internal support tool for managing service-level commitments and workflows. The body was unavailable, so its implementation details and practical takeaways cannot be verified.

The post is titled “Packaging Supabase with Nix.” Its body was unavailable, so the technical approach and takeaways cannot be verified from the title and URL alone.

Raminder Singh explains PostgreSQL’s roles, privileges, ownership, group membership, grant options, and default access privileges through hands-on psql examples. The guide highlights subtle permission behavior, including inherited privileges from the public role and practical patterns for managing database access securely.

The post appears to explain how to build semantic image search using Amazon Titan embeddings and Supabase Vector. The body was unavailable, so this summary is based on the title and URL alone.

Supabase explains how it evolved from manual to automated performance testing for WebSocket services. The post covers k6 load generation, Grafana and Prometheus observability, Telegraf metric buffering, Terraform-based infrastructure, and CI integration, with practical setup guidance and scaling results.

Supabase engineers explain how Matryoshka Representation Learning enables OpenAI embeddings to be truncated while preserving semantic information. They benchmark Adaptive Retrieval with pgvector, showing how a low-dimensional HNSW first pass followed by full-vector re-ranking can reach 99% accuracy at up to 580 queries per second.

The post announces MongoDB compatibility for Supabase projects through FerretDB, allowing applications to use a MongoDB-compatible interface backed by Postgres. The body was unavailable, so this summary is based on the title alone.

The post announces pgvector 0.6.0, highlighting parallel index builds and a claimed 30x speed improvement. The body was unavailable, so this summary is based on the title and URL alone.