PostgreSQL 19 is adding query plan hints, called “advice.” Before overriding the planner: • Run ANALYZE • Check indexes and fix statistics. If you still need advice, scope it by query ID and keep checking EXPLAIN. See how pg_plan_advice works:

postgres experts: It's unfortunate that pg-jev won't be available in @supabase (not in the current form) Right now pg-jev depends on plpython3u for caching and concurrency, which is classfied as untrusted language Rewriting it in plpgsql and http would cause the loss of concurrency. A call on 2000 rows would talke about 8,2 seconds instead of 1,7. Can't think of an alternate way right now, so I'd appreciate any help

Postgres 19 is one of those releases you feel more than you notice. Built-in repack, better replication, and steady improvements across the board. Dive into this breakdown from @craigkerstiens on what actually matters:

Postgres compatibility is a start. Production is the real test. For AI apps, that means scale, availability, recovery, security and fewer brittle pipelines between transactional and analytical data. Check out this breakdown on Snowflake Postgres for teams building beyond the demo:

Introducing PostgreSQL for agents in AlloyDB! Agent queries can overwhelm standard databases. Now you can: - Keep core business systems protected - Get fast access to live operational data - Scale to 1,000+ readers on demand Read how it works 👉

Introducing PostgreSQL Hub. Explore resources, guidance, and tools to help organizations build, optimize, and scale PostgreSQL workloads on Azure. Learn more:

Scale PostgreSQL reads the easy way. In #AzureFriday#, @shanselman and Paula Berenguel show how PostgreSQL on Azure uses read replicas and virtual endpoints to handle heavy read workloads, just like OpenAI powers ChatGPT on Postgres. Watch now: