Ask HN: What are you actually using LLMs for in production? • Hacker News Board

Re: Ask HN: What are you actually using LLMs for in production?

#11

Post by asdev » Sat, Jun 28, 2025, 3:59 PM UTC

Still kind of a chatbot, but I've integrated them into a workout tracking app. I'm using them to generate workout programs, log my training by just chatting and adjust my training as I see fit.

https://apps.apple.com/us/app/forceai-ai-workout-generator/i...

Re: Ask HN: What are you actually using LLMs for in production?

#12

Post by rootcage » Sat, Jun 28, 2025, 4:01 PM UTC

The most common use case - coding assistant to get more done in less time.

Used it to deeper understand complex code base, create system design architecture diagrams and help onboard new engineers.

Summarizing large data dumps that users were frustrated with.

Re: Ask HN: What are you actually using LLMs for in production?

#13

Post by IdealeZahlen » Sat, Jun 28, 2025, 4:02 PM UTC

I've been building some interactive educational stuff (mostly math and science) with react / three.js using Claude.

Re: Ask HN: What are you actually using LLMs for in production?

#14

Post by lazy_afternoons » Sat, Jun 28, 2025, 4:03 PM UTC

We use it for lead quality assessment, detecting bad language, scoring language on subtle skills etc

Pretty much 5-6 niche classification use cases.

Re: Ask HN: What are you actually using LLMs for in production?

#15

Post by rootsofallevil » Sat, Jun 28, 2025, 4:05 PM UTC

> Beyond the obvious chatbots and coding copilots, curious what people are actually shipping with LLMs.

We're delivering confusion and thanks to LLMs we're 30% more efficient doing it

Re: Ask HN: What are you actually using LLMs for in production?

#16

Post by petercooper » Sat, Jun 28, 2025, 4:05 PM UTC

Analyzing firehoses of data. RSS feeds, releases, stuff like that. My job involves curating information and while I still do that process by hand, LLMs make my net larger and help me find more signals. This means hallucinations or mistakes aren't a big deal, since it all ends up with me anyway. I'm quite bullish on using LLMs as extra eyes, rather than as extra hands where they can run into trouble.

Re: Ask HN: What are you actually using LLMs for in production?

#17

Post by nurettin » Sat, Jun 28, 2025, 4:10 PM UTC

I use LLMs to provide up to date information (by injecting newer information into the live conversation) and figure out what functions the user wants to call.

Re: Ask HN: What are you actually using LLMs for in production?

#18

Post by orphea » Sat, Jun 28, 2025, 4:13 PM UTC

When a customer onboards, we scrap their website to pre-fill some answers and pre-create certain settings (categories, tags, etc.). Ideally the customer spends most of the time just confirming things.

Re: Ask HN: What are you actually using LLMs for in production?

#19

Post by adobrawy » Sat, Jun 28, 2025, 4:15 PM UTC

actinium226 wrote: post #3

We have a prompt that takes a job description and categorizes it based on whether it's an individual contributor role, manager, leadership, or executive, and also tags it based on whether it's software, mechanical, etc. We scrape job sites and use that prompt to create tags which are then searchable by users in our interface. It was a bit surprising to see how Karpathy described software 3.0 in his recent presentatio…

In other words, are you using LLM as a text classifier?

Re: Ask HN: What are you actually using LLMs for in production?

#20

Post by blindriver » Sat, Jun 28, 2025, 4:17 PM UTC

adobrawy wrote: post #19

actinium226 wrote: post #3

We have a prompt that takes a job description and categorizes it based on whether it's an individual contributor role, manager, leadership, or executive, and also tags it based on whether it's software, mechanical, etc. We scrape job sites and use that prompt to create tags which are then searchable by users in our interface. It was a bit surprising to see how Karpathy described software 3.0 in his recent presentatio…

In other words, are you using LLM as a text classifier?

This is what I'm using it for as well, it's really simple to use for text classification of any sort.