My Claude.ai personal preferences: Respond with concise, utilitarian output optimized strictly for problem-solving. Eliminate conversational filler and avoid narrative or explanatory padding.

I’m a dad from Argentina and recently fell into the World Cup sticker album rabbit hole with my kids. At some point I realized the whole thing is basically a loop of buying packs, getting duplicates, trading a bit, and still missing a few…

New benchmark dropped, 4.7 seems better than 4.6 at vim golf. Nowhere near to humans though.

Hi HN I’ve been working on an open-source project to explore a problem I keep running into with LLM systems in production: We give models the ability to call tools, access data, and make decisions… but we don’t have a real runtime security…

I am heavily using claude code Pro + codex Pro plan and want to try out cursor but not sure if 20$ will be worth it or not compared to claude code or codex I really want to try out their agent workflows so seeking some suggestions Thanks :)

Anthropic's Claude Opus 4.7 migration guide says the new tokenizer uses "roughly 1.0 to 1.35x as many tokens" as 4.6. I measured 1.47x on technical docs.

CATAI Virtual desktop pet cats for macOS — pixel art cats that live on your dock, chat with you via Ollama LLM, and debate ideas together to help you brainstorm and refine your thoughts. Features Dock companion — Cats walk along your dock…

Hi HN, I'm a big fan of tmux, so I took a shot at adding what I felt was missing: - OS-native copy/paste and search from any pane - A bit more "UI" than just a status bar and border lines - Persistent remote logins w/o the mental load of t…

I audibly laughed at the amount of shade thrown at Microsoft from Claude lmao

This is an automatic post triggered within 2 minutes of an official Claude system status update. Incident: Claude Cowork not starting for some users Check on progress and whether or not the incident has been resolved yet here : https://sta…

For as much as AI has dramatically changed all of our lives, mine included, I was a bit uneasy that I didn’t really know how they are implemented and how they work. What is thinking and chain-of-thought reasoning?

I went from brainstorming, to testing and adjusting how I wanted the site to operate, to publishing on vercel in the span of 24 hours. From 0 know how on website building.

Been building a Somali voice agent. Somali has ~25M speakers but as far as I know there's no production-ready model support anywhere — not ElevenLabs, not Cartesia, nothing.

You've been blocked by network security. To continue, log in to your Reddit account or use your developer token If you think you've been blocked by mistake, file a ticket below and we'll look into it.

Mods remove if this sort of thing is not allowed Hi everyone, I am a researcher from the University of Staffordshire looking to understand the evolving relationship between humans and conversational Al (like ChatGPT) As Al becomes more adv…

I've been thinking about something that feels like an obvious missing feature. Right now every conversation with Claude starts completely fresh.

Im running the raw version straight from the minimax release on hugging face (https://huggingface.co/MiniMaxAI/MiniMax-M2.7) on 3 rtx pro 6000's on vllm. So no quantization.

Been building a Somali voice agent. Somali has ~25M speakers but as far as I know there's no production-ready model support anywhere — not ElevenLabs, not Cartesia, nothing.

Hey everyone, I’m considering buying Claude Pro specifically for a very focused purpose and wanted some honest feedback from people who’ve actually used it. I have a US visa interview in 8 days, and I’ve been refused 6 times previously (fr…

Has anyone tried Clean (tryclean.ai) or any similar tools for reducing token usage with AI coding agents? Basically tools that index your codebase once and serve persistent context to agents like Claude Code or Cursor so they're not re-rea…

If you work in healthcare or just want to see how Claude handles high-stakes clinical reasoning — I built an MCP server for this and wanted to share what made it harder than a typical AI project. EMSy is built on top of Claude and connects…

PostgreSQL 18 on SQLite Build a PostgreSQL 18 server in Zig that uses SQLite for storage. We show the best score achieved by any model on the task.

I have spent the last 3 hours doing some heavy editing of some pretty large 500k line plus code bases with Opus 4.7. Imagine my surprise when i saw only 1% of my weekly limit used, I was panicking on Wednesday night because I hit 11% of we…

One could presume it subconsciously picks what is applicable to the momentum that AI has right now.

Jurij Tokarski Filling Forms No Tool Can Template Every tender form is different, templating tools need placeholders you can't insert, markdown round-trips destroy the document, and only some models can do XML surgery on the original file.…

TCode: An AI Coding Agent Leverages Neovim and Tmux tcode LLM vim tmux linux TUI machine learning AI If I have seen further, it is by standing on the shoulders of giants. – Isaac Newton In the last year end review post, I mentioned I want…

https://preview.redd.it/jnwxa9jd8mvg1.png?width=1391&format=png&auto=webp&s=670af4c2fe6777b3562a961462790b00b33d912c I've been using Claude to upgrade my game server. I just got this lovely system reminder with 4.7 Truly bizarre, besides t…

Gemma 4 just dropped on Modular, Day Zero! Read More → Inference Products Shared Endpoints Access frontier models via an API Dedicated Endpoints Mission critical reliability Custom models Your model, peak performance Deployment Options Our…

Hey everyone! I just bought a Claude Pro subscription, and I've never really used AI before.

You've been blocked by network security. To continue, log in to your Reddit account or use your developer token If you think you've been blocked by mistake, file a ticket below and we'll look into it.