Index only the rows that queries ask for, compare the size with a full index, and enforce a unique rule for active rows. After this, you know why a prepared statement can ignore a partial index, and how to avoid it.
Add a GIN index to a JSONB column, filter with the contains operator, and compare it with an expression index and jsonb_path_ops. After this, you can tell which JSONB filters use an index and which scan the table.
Read pg_stat_user_indexes to find indexes that were never scanned, find identical indexes in pg_index, and drop them with DROP INDEX CONCURRENTLY. After this, you can remove indexes that cost writes and give nothing back.
Reproduce a deadlock between two money transfers, read the message, and fix it by locking rows in the same order. After this, you can turn a deadlock detected error into a plain wait.
Let two workers take different jobs from one table with FOR UPDATE SKIP LOCKED, and see what happens to a job when a worker fails. After this, you can use a plain table as a safe queue.
Add a NOT NULL column to a big table with a constant default, or with a nullable column, a batch backfill, NOT VALID and VALIDATE. After this, you can tell a quick change from a table rewrite and avoid long locks.
Walk down an employee tree, walk up to the top, count each manager's team and guard the query against a loop with CYCLE. After this, you can query a parent-child table of any depth.
Take a pg_dump backup in Docker, restore it into a scratch database, compare the tables and see what a damaged backup looks like. After this, you have a short routine that proves your backup works.