beklein

Relevant blog post from simonw: https://simonwillison.net/2026/Feb/7/software-factory/

https://www.spacex.com/updates#xai-joins-spacex additionally the longer article on SpaceX site

This will actually work well with my current workflow: dictation for prompts, parallel execution, and working on multiple bigger and smaller projects so waiting times while Codex is coding are fully utilized, plus easy commits with auto commit messages. Wow, thank you for this. Since skills are now first class tools, I will give it a try and see what I can accomplish with them.

I know/hope some OpenAI people are lurking in the comments and perhaps they will implement this, or at least consider it, but I would love to be able to use @ to add files via voice input as if I had typed it. So when I say "change the thingy at route slash to slash somewhere slash page dot tsx", I will get the same prompt as if I had typed it on my keyboard, including the file pill UI element shown in the input box. Same for slash commands. Voice is a great input modality, please make it a first class input. You are 90% there, this way I don't need my dictation app (Handy, highly recommended) anymore.

Also, I see myself using the built in console often to ls, cat, and rg to still follow old patterns, and I would love to pin the console to a specific side of the screen instead of having it at the bottom and pls support terminal tabs or I need to learn tmux.

Thank you! Learned something new today. Will try to look out for this trick on other Lovable sites I will stumble upon.

Would love to see the original prompt for Nano Banana from OP somewhere. One that yields decent results, for me, is:

{ "image_generation_prompt": { "subject_focus": { "primary": "Architectural exterior scene", "constraint": "Strictly preserve original building geometry, facade details, and structural layout", "reference_adherence": "High structural fidelity to input image" }, "environment_and_season": { "season": "Late November, very late autumn", "weather": "Post-rain, overcast, gloomy, high humidity", "sky": "Heavy grey cloud cover, diffuse white/grey light, no direct sunlight", "ground_texture": "Wet asphalt/pavement, highly reflective puddles, wet concrete, scattering of wet brown decaying leaves" }, "vegetation_details": { "trees": "Leafless branches, dormant skeletal trees, sparse lingering brown foliage", "color_palette": "Desaturated greens, browns, greys, russet, damp earth tones", "state": "Winter-ready, wet bark, dormant landscaping" }, "human_element": { "density": "Sparse, minimal crowd", "clothing": "Heavy winter coats, scarves, boots, muted colors", "activity": "Walking briskly to avoid cold, holding closed wet umbrellas, hurrying, heads down against the wind", "mood": "Solitary, cold, urban transit" }, "photographic_style": { "medium": "Realistic architectural photography", "camera": "35mm lens, sharp focus on architecture", "tone": "Cinematic, moody, desaturated, cool color temperature, blue-grey tint", "quality": "8k resolution, high dynamic range, hyper-realistic textures" } } }

As far as I can tell, they say: "Mission control and data distribution are managed by EUMETSAT." They have published their own blog post here: https://www.eumetsat.int/features/see-earths-atmosphere-neve...

There they say that: "Observations made by MTG-S1 will feed into data products that support national weather services …". So I guess there will be no simple, publicly available REST API or so... but if anybody finds anything, let us know here :)

Wasn't aware you're so active on HN; sorry for stealing your karma.

Love the idea of a dedicated series/pod where normal people take on hard problems by using and leveraging the emergent capabilities of frontier AI systems.

Anyway, thanks for pod!

The Latent Space podcast just released a relevant episode today where they interviewed Kevin Weil and Victor Powell from, now, OpenAI, with some demos, background and context, and a Q&A. The YouTube link is here: https://www.youtube.com/watch?v=W2cBTVr8nxU

I wonder if electrical engineers felt this way about the reliability of the silicon crystal lattice their circuits rely upon…

Anthropic posted an AMA style interview with Amanda Askell, the primary author of this document, recently on their YouTube channel. It gives a bit of context about some of the decisions and reasoning behind the constitution: https://www.youtube.com/watch?v=I9aGC6Ui3eE

Not the author/contributor, but the app is built using Tauri for easy multi-platform support, so the backend logic is implemented in Rust and the frontend UI is implemented in TypeScript. I think it’s a valid choice. GitHub does not include any model _code_ in the stats; the models will be downloaded separately the first time you use them. Hope this helps.

I know many people hate sites like this, but I actually like them for these use cases. You can get a quick, LLM-generated overview of the architecture, e.g. here: https://codewiki.google/github.com/cjpais/handy

Most of my complex documents are, luckily, Markdown files.

I can recommend https://github.com/tobi/qmd/ . It’s a simple CLI tool for searching in these kinds of files. My previous workflow was based on fzf, but this tool gives better results and enables even more fuzzy queries. I don’t use it for code, though.

Thanks for the post and the explanation.

I really enjoyed this relevant article about prompt caching where the author explained some of the same principles and used some additional visuals, though the main point there was why KV cache hits makes your LLM API usage much cheaper: https://ngrok.com/blog/prompt-caching/

Not all laws are hard science laws in the sense of the second law of thermodynamics; they are, however, good approximations based on experience and, in the right context, make a bit more sense.

Notes from the original author:

> I've been involved in spacecraft and space systems design and development for my entire career, including teaching the senior-level capstone spacecraft design course, for ten years at MIT and now at the University of Maryland for more than a decade. These are some bits of wisdom that I have gleaned during that time, some by picking up on the experience of others, but mostly by screwing up myself. I originally wrote these up and handed them out to my senior design class, as a strong hint on how best to survive my design experience. Months later, I get a phone call from a friend in California complimenting me on the Laws, which he saw on a "joke-of-the-day" listserve. Since then, I'm aware of half a dozen sites around the world that present various editions of the Laws, and even one site which has converted them to the Laws of Certified Public Accounting. (Don't ask...) Anyone is welcome to link to these, use them, post them, send me suggestions of additional laws, but I do maintain that this is the canonical set of Akin's Laws...

Correct, check the "Our name" section at: https://n8n.io/press/

Thanks for the Google link, I was just asking GPT-5-Pro "How to Google: 'How do I run a rust project'", and am still waiting for the answer... the point was that searching for an answer (wherever/however) is not necessary in some cases, like this one, but asking the AI agent to find a solution can be sufficient and is totally ok. Engineers are allowed to delegate, there is not nothing wrong with this.

If you've never used Rust before, I couldn't find good documentation on how to run a existing Rust project nor could I find `cargo install` on the "Getting Started" page. I could read the Cargo Book, or check `--help` I guess, but this can be surprisingly time consuming as well, it might take 5-30 minutes of active searching to locate the information. If you can, try put yourself in a beginner's mindset and think though your argument again.

Regarding your second point, I think people actually underutilise LLMs for simple tasks. Delegating these tasks frees up your problem-solving skills for challenges that truly need human insight. In this case, asking an LLM is arguably the smart choice: it's a common task in training data, easy to verify, and low-risk to run and not a direct learning or benefit for your initial question.

I love OpenRouter, since it is a simple way to get started and provides a wide range of available models.

You can buy credits and set usage limits for safe testing per API key to gain access from many AI models through one simple and unified API from all popular model providers (OpenAI, Anthropic, Google, xAI, DeepSeek, Z.AI, Qwen, ...)

Ten dollars is plenty to get started... experiments like in the post will cost you cents, not dollars.

Comparing this blog post to a 500-page book or a multi-hour course and calling it “weaksauce” misses the point. This post is meant as an introduction to the dot product, and it does that really well. The formal definition (6.1) and explanation in Axler’s book wouldn’t make a good starting point for most people, it isn't even a good next step in my opinion. It’s great that you’re passionate about the topic, really, but helping more people discover math means meeting them where they are and appreciating content like this for what it’s trying to do.

It is a personal page, see the About page: https://european.cloud/about/

I agree with you that this distinction could be made more prominently.

claude --model claude-haiku-4-5-20251001

You can use the model flag and specify the model like: claude --model claude-haiku-4-5-20251001

Here a clip of that scene: https://www.youtube.com/watch?v=7bXJ_obaiYQ

When writing documentation, you need to establish a baseline of required knowledge and skills for your audience. You can choose any level, but deviating too far above or below that baseline will inevitably frustrate some readers.

When this happens, you can either make excuses or focus on solutions. Problems can be difficult, but with modern tools like AI systems, Google, or even books, it has never been easier to overcome them. If you don’t know what a Shoobababoo is or why you should use the quagmire instead of the hoobastank, you can look it up. Ideally, documentation would contain every answer with minimal need for external knowledge transfer, but the world doesn’t owe you that convenience.

Relevant blog post: https://ghuntley.com/cursed/

Technical Report: https://arxiv.org/abs/2508.18255

Download Models: https://huggingface.co/collections/NousResearch/hermes-4-col...

It works fine in OpenRouter

Again, totally fair points, and for many people a (simple) GUI is the right tool. I am not against GUIs or particularly pro LLMs; I just want to show an alternative way to solve video editing problems without judging any specifc technology.

Not all video work is visual-first storytelling. In engineering/lab contexts you often just need “good enough” trims, concatenation, speedups, and a few labels to document an experiment. As I said in another comment, I usually get the timestamps by noting them down while watching the video, or in rarer cases from timestamped sensor data.

Sorry if my explanation wasn't good enough...

In my common use case, I note down the timestamps of the cuts. No LLM needed here.

I know there are options for everything in FFmpeg, and I’m thankful to the community and maintainers for providing such a powerful and well-documented tool. I sometimes just want a quick video edit that doesn’t involve reading the man pages for minutes or hours.

I’m not saying this is better than a GUI, but instead let me give you an example.

Sometimes I have movie clips A, B, and C. I want to trim part of A at the beginning and end, then stitch clip B to it while speeding it up by 2x, and finally add clip C at the end, also trimmed a bit. After that, I want to add some text at specific times for a specific duration. In the end, I’ll export everything in 1080p.

I know all of this can be done with Final Cut or any other video editing app using a GUI, or the cool Emacs tool above...

But for me this would mean (GUI example) downloading the app and watching YouTube tutorials just to learn what to click.

For simple video editing (and other tasks), I sometimes need to get the job done quickly without wanting to learn a whole new tool.

I found out that I can achieve sufficiently advanced video edits with FFmpeg commands produced step by step from a LLM.

If you think this is awful, ok. I thought it was neat and wanted to share the idea.