Supabase Compute (supabase.com)

Supabase has introduced Supabase Compute in private alpha, enabling developers to run agents, background jobs, and APIs directly alongside their databases within a single project. This new feature addresses limitations of Edge Functions by allowing resource-intensive tasks that require more memory or CPU, and systems that run longer than 400 seconds. Available initially in the us-west-2 region, Supabase Compute supports various languages via Docker, empowering developers to manage tasks like AI agents, durable job processors, and real-time WebSocket applications seamlessly within their existing Supabase environments. The significance of Supabase Compute lies in its potential to simplify development workflows by consolidating various tasks into a single platform, eliminating the need for multiple deployment environments. With each service operating in a full Linux environment, developers can optimize performance, as exemplified by Dreambase's data agent, which achieved median query speeds of just 2.9 ms compared to 65 ms from another region. Though it's currently in alpha and not suitable for production use, developers can join a waitlist to explore the feature. Key functionality includes customized service deployment through the Supabase CLI, though limitations such as lack of autoscaling and persistent disk must be noted.