[simonw]

So run the LLM in an agent loop: give it a benchmarking tool, let it edit the configuration and tell it to tweak the settings and measure and see how much if a performance improvement it can get.

That's what you'd do by hand if you were optimizing, so save some time and point Claude Code or Codex CLI or GitHub Copilot at it and see what happens.

[IgorPartola]

“We will take all the strokes off Jerry's game when we kill him.” - the LLM, probably.

Just like Mr Meeseeks, it’s only a matter of time before it realizes that deleting all the data will make the DB lightning fast.

[simonw]

Exactly true, which is why you need to run your agent against a safe environment. That's a skill that's worth developing.

[lomase]

How much that would cost?

[simonw]

Probably about 10 cents, if you're even paying for tokens. Plenty of these tools have generous free tiers or allowances included in your subscription.

I run a pricing calculator here - for 50,000 input tokens, 5,000 output tokens (which I estimate would be about right for a PostgreSQL optimization loop) GPT-5 would cost 11.25 cents: https://www.llm-prices.com/#it=50000&ot=5000&ic=1.25&oc=10

I use Codex CLI with my $20/month ChatGPT account and so far I've not hit the limit with it despite running things like this multiple times a day.

[lomase]

If optimizing a Postgress SQL server cost 11.25 cents and everybody can do it because AI how much are you going to bill your customer? .20 cents?

If that is true in some months there will be no dba jobs.

Funny that at the same time SQL is one of the most requested languages in job postings.

[simonw]

Knowing what "optimizing a PostgreSQL server's configuration" even means continues to be high value technical knowledge.

Knowing how to "run an agentic loop to optimize the config file" is meaningless techno-jabber to 99.99% of the world's population.

I am entirely unconcerned for my future career prospects.

[lomase]

So your big advantage is that nobody has lauched agentic tools for the end user yet?

[simonw]

Anyone can learn to unblock a sink by watching YouTube videos these days, and yet most people still hire a professional to do it for them.

I don't think end users want to "optimize their PostgreSQL servers" even if they DID know that's a thing they can do. They want to hire experts who know how to make "that tech stuff" work.

[lillecarl]

They charge per token...

[danielbln]

Not if you're on a subscription they don't.

[lillecarl]

So the hidden usage caps doesn't equate to token usage?

They charge per token, everyone charges per token.