Supabase and AWS Help Developers Build Fast and Scale to Millions

Serving 5 million builders worldwide, Supabase unveils Amazon S3 integrations to take away technical hurdles as apps develop

Early-stage tasks that begin as weekend experiments can evolve shortly into enterprise-grade apps

At AWS re:Invent, Amazon Web Services, Inc. (AWS), an Amazon.com, Inc. firm (NASDAQ: AMZN), and Supabase, the Postgres improvement platform, in the present day introduced two new Amazon Simple Storage Service (Amazon S3)-based storage improvements and a brand new ETL¹ characteristic that make constructing generative synthetic intelligence (AI) brokers and apps simpler. Built on Apache Iceberg and Amazon S3 Tables, Supabase Analytics Buckets assist analytics workloads, whereas Supabase Vector Buckets present specialised storage that powers AI options like semantic search and personalization. Supabase ETL robotically strikes information from Postgres databases to analytics instruments with a single click on, eliminating months of coding work. Built on AWS, Supabase has launched greater than 10 million databases to date and has change into the inspiration of selection for startups, with over 60% of every Y Combinator batch constructing on the platform.

These instruments assist builders construct apps that customers love and companies want. Customers can scale apps seamlessly from prototype to manufacturing methods, serving tens of millions of customers with out costly rebuilds that decelerate rising corporations. Supabase handles all of the behind-the-scenes work that AI code technology instruments want to create totally useful apps, with PostgreSQL, one of many world’s most generally used databases, as a single level of management. The platform, which serves 5 million builders worldwide and runs on AWS, has change into a key enabler of the vibe coding motion, the place builders keep in a artistic circulation state whereas AI instruments deal with the complexity of constructing production-ready purposes.

“Before Supabase, constructing an app meant juggling a number of separate companies—one in your database, one other for consumer logins, a 3rd for file storage—every with its personal dashboard and method of working,” stated Paul Copplestone, CEO and co-founder, Supabase. “Today, Supabase brings all of those collectively in a single platform, all constructed on prime of Postgres. This means builders work in a single place as an alternative of 5, with the boldness that AWS’s scale will deal with the whole lot from their first consumer to their millionth with out lacking a beat.”

Supabase at present operates throughout 17 international AWS Regions, together with Asia Pacific (Singapore), Asia Pacific (Tokyo), Asia Pacific (Sydney), Europe (London), and US West (Northern California), enabling builders to create databases nearer to clients for quicker response instances. This means an app responds immediately whether or not a consumer in Tokyo searches for a product or a gamer in Sydney joins a match—AWS’s international infrastructure delivers the split-second efficiency that makes apps really feel seamless and responsive. Supabase additionally runs solely on AWS Graviton processors, delivering improved efficiency with decrease operational prices.

Key particulars of in the present day’s launch embrace:

• Supabase ETL robotically strikes information from a Postgres database to a unified information layer that powers each analytics and AI options. With a single click on, ETL copies information to each Supabase Analytics Buckets and Supabase Vector Buckets, giving dashboards and AI purposes clear, organized information.
• Supabase Analytics Buckets assist the Apache Iceberg format on Amazon S3 Tables, which implies analytical information will get saved in a format that Amazon and third-party companies can learn natively. When a buyer needs to run dashboards or reviews, Supabase ETL replicates information from a consumer’s important Postgres database into an Analytics Bucket. Customers can then question information from Amazon Athena, Amazon Redshift, Amazon EMR, or Amazon Quick Sight with out placing load on manufacturing databases.
• Supabase Vector Buckets let customers retailer giant vector datasets in Amazon S3 as an alternative of Postgres databases. This issues for options like suggestion engines and semantic search. When a buyer searches for “summer season clothes” in a buying app, conventional search seems for precise phrase matches, whereas vector search understands ideas and can discover associated gadgets even when they use completely different phrases (like “sundress” or “heat climate outfits”). You nonetheless question Vector Buckets from Postgres utilizing the identical interface as earlier than, however the storage sits in S3 the place it’s extra cost-efficient and can scale to tens of millions of embeddings with out stressing your database.

This structure makes use of PostgreSQL because the core transactional database for reside enterprise operations like processing orders. Supabase ETL constantly replicates information to Supabase Analytics Buckets for historic reporting and enterprise intelligence whereas Supabase Vector Buckets deal with AI-powered good suggestions and semantic search. Everything syncs in close to real-time, so an e-commerce firm can write one question to present a buyer their present order (PostgreSQL transactional information), analyze their shopping for historical past (Supabase Analytics Buckets), and counsel customized merchandise (Supabase Vector Buckets) – all from a single interface as an alternative of three separate methods.

“Every fashionable enterprise is a knowledge enterprise and Amazon S3 is foundational for builders,” stated Mai-Lan Tomsen Bukovec, vp of Technology, AWS. “By bringing collectively S3’s scale and reliability with Supabase’s built-in platform, we’re making it simpler for builders to work with their information and transfer from AI experimentation to purposes in manufacturing.”

“Imagine a retailer attempting to analyze buyer habits throughout their web site, cell app, and bodily shops. They’d want to gather information from a number of methods, clear it up, translate it into a typical language, and ship it to the place analysts can use it — all whereas preserving it up to date in close to real-time,” added Copplestone. “What we’ve finished with AWS is flip this whole course of into one thing so simple as ticking a field, permitting companies to deal with utilizing their information somewhat than struggling to entry it.”

In the third quarter of 2025 alone, extra tasks had been created on Supabase than within the first 4 years of the corporate mixed. Startups like Lovable, Figma Make, and Bolt depend on Supabase to scale quickly on AWS. Lovable, an AI web site builder, makes use of Supabase to autonomously spin up databases every time customers create new purposes — showcasing the platform’s potential to energy agentic workloads at scale.

The submit Supabase and AWS Help Developers Build Fast and Scale to Millions first appeared on AI-Tech Park.