frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Ask HN: Any advice on how to learn good software architecture practices?

5•jimsojim•1h ago•2 comments

Ask HK: How are you building AI apps today?

4•Mnexium•3h ago•3 comments

Ask HN: What Is an "AI Engineer"?

16•seattle_spring•1d ago•27 comments

Ask HN: Does Claude Code remove the need for so many front-end frameworks?

7•ex-aws-dude•20h ago•10 comments

Ask HN: Are Tech Meetups Dead?

35•mlinhares•5d ago•29 comments

A disk-first C++ vector engine

4•saeedq•17h ago•0 comments

Ask HN: I found out that I'm about to be laid off. How do people find jobs?

20•wwwthrowaway256•23h ago•15 comments

Garnix, the Nix CI, is shutting down

12•cinericius•1d ago•4 comments

Do not use Cloudflare DNS regsitrar

10•talkingtab•19h ago•0 comments

Ask HN: How do you feel about posts about GenAI taking over the HN front page?

11•blenderob•1d ago•15 comments

Train 1T parameter LLM with 8 GPUs?

3•kendy1992•1d ago•1 comments

Ask HN: Why not have an EU browser?

6•osigurdson•1d ago•20 comments

Ask HN: Is anyone working at least 4 hours daily on an Apple Vision Pro?

152•widenrun•3d ago•108 comments

Sqlit – A lazygit-style TUI for SQL databases

20•MaxTeabag•3d ago•5 comments

Ask HN: How do you model temporarily invalid data structures

5•escot•2d ago•1 comments

Did the Linux memory management maintainer "just quit"?

17•hliyan•3d ago•3 comments

Ask HN: When and why did you start believing in God?

27•dvrp•3d ago•87 comments

Ask HN: Why didn't the C64 come with Simons' BASIC in the box from 1983 onward?

15•amichail•4d ago•14 comments

Ask HN: Did Messages get removed from Google Takeout?

22•MollyRealized•5d ago•1 comments

Amdahl's Law for LLM generated code

5•akiarie•3d ago•4 comments

You've reached the end!

Open in hackernews

Ask HK: How are you building AI apps today?

4•Mnexium•3h ago
Generally curious how folks are using LLM APIs today.

Are folks doing anything bespoke, unique or different?

I think we've seen a lot of usage with harnesses, CLIs, Cursor/Windsurf etc which uses the APIs underneath.

If you are using the API itself, what does your architecture look like with said API?

Comments

late_night_fix•2h ago
I try to batch requests and reuse responses wherever possible.LLM API calls are awesome,but if you don't watch your usage,cost sneaks up fast.
Mnexium•2h ago
Agreed, are you doing anything to take advantage of cacheing?

Anything else you can share about project?

LogicCraft678•2h ago
Mostly multi-model routing and structured outputs