csoham
Re:Tokenminning: Because Tokenmaxxing Is a Bad Idea
on Tuesday June 09, 2026 @09:04AM (#48460595)
Cool stuff! I suspect we will now see a bunch of startups coming up focused on helping companies reduce their AI usage, similar to how we have companies focused on optimizing cloud costs.
A Better Practices Guide to Using Claude Code
The PSF has withdrawn $1.5M proposal to US Government grant program
Re:Man lands in hospital after Samsung smart ring battery swells and t...
on Tuesday September 30, 2025 @05:50AM (#45423683)
Wow this is crazy. What's crazier is the drop in battery level from 7 days to 1.5 days in 9 months!
Agentic Commerce Protocol Spec
Developing an open standard for agentic commerce
Re:Ask HN: What are you working on? (September 2025)
on Tuesday September 30, 2025 @03:08AM (#45422749)
I'm working on ScaleDown [1], a context pruning API.
So over the past few years, I have seen how contexts have been steadily growing in AI apps. And while the context lengths of LLMs have also been increasing, they are still effectively about 200k tokens. The performance drops off a cliff after that (you might have noticed it as well with long AI chats).
It is a simple API that prunes away irrelevant parts of a context for a given prompt, a.k.a. context-aware pruning. Integration is super simple: just an extra API call before the final LLM API call. You can get an API from the website.
I would love to chat if this is something that is relevant to you and if you have any feedback on what we are building!
Re:When the job search becomes impossible
on Wednesday September 17, 2025 @10:06AM (#45276015)
[self promotion alert] the "you are not alone" point really resonated with me. When I lost my job, I was alone, helpless and not sure what the next steps were. This is why I tried to create a community of people willing to support and be a listening ear for people going through job loss and this tough job market. It's at layoff.supprt. honestly I have not been supporting it for a while but of you find this helpful and would like more features then do let me know!
Re:Tau² Benchmark: How a Prompt Rewrite Boosted GPT-5-Mini by 22%
on Wednesday September 17, 2025 @09:59AM (#45275945)
Really intresting. What did the original prompt look like? Perhaps the original prompt was not that good? I feel like the changes claude suggested (except a couple maybe) are already pretty well known prompt engineering practices.
Re:VLLM: Anatomy of a High-Throughput LLM Inference System
on Monday September 01, 2025 @04:16PM (#45096244)
One of the best deep-dives I have seen!
A Novel Pretrained Tokenizer-Free LLM Architecture
AI API Prices are 90% Subsidized
Re:Ask HN: What are you working on? (April 2025)
on Tuesday April 29, 2025 @02:51PM (#43836555)
https://extension.scaledown.ai
I realised that I'm a horrible prompt writer and so are many other people. So i created this extension to help me write better prompts, using templates, save prompts, and optimize it for better performance. No login, not paid, just a useful little extension. Check it out!
Re:Show HN: laidoff.support Support for laid off workers
on Saturday February 03, 2024 @12:09PM (#39242342)
I created this website because being laid off is one of the worst feelings ever, and it is hard to find supportive people to help you out.
All resources are free or very cheap.
Help support your fellow laid-off workers by joining as a mentor!