Hacker News Reader

ukuinaon Datastar: Web Framework for the Future?

There's no real comparison between Python/Node and Golang's cross-compilation.

ukuinaon Roo or Cline? We're building a superset

Have you tried https://github.com/block/goose ?

ukuinaon Open source and self hostable/private file converter

pandoc-wasm?

ukuinaon OpenAI gets ready to launch GPT-4.1

> OpenAI is getting ready to unveil a number of new AI models, sources familiar with the company’s plans tell The Verge. Among the new AI models will be a release of what I’m expecting will be branded GPT-4.1, which one source describes as a revamped version of OpenAI’s GPT-4o multimodal model.

> OpenAI is also readying the full version of its o3 reasoning model and an o4 mini version that could debut even sooner. AI engineer Tibor Blaho discovered references to o4 mini, o4 mini high, and o3 in a new ChatGPT web version earlier today, suggesting these additions are imminent. I understand o3 and o4 mini are both set to debut next week, unless OpenAI moves the launch plans around.

ukuinaon Ask HN: What's the best way to get started with LLM-assisted programing?

> Prompt engineering is [...] a form of spellweaving where you have to find the edges of all relevant forms of existence and non-existence, codify them into imperative language, add the right nuance to finesse the style and have the confidence to cast the spell — over and over and over again.

This brought a tear to my eye.

ukuinaon Ask HN: What are you working on? (March 2025)

> I'm almost certainly treading ground well explored by others

This is a good thing. One requires smoothed pathways to run!

ukuinaon Launch HN: Continue (YC S23) – Create custom AI code assistants

> The recent productivity has been so high, it has the effect of making my previous efforts seem sort of pedestrian

Fantastic!

Agentic task-offloading is already here; it is just unevenly distributed.

ukuinaon I genuinely don't understand why some people are still bullish about LLMs

ChatGPT Advanced Voice Mode?

ukuinaon I genuinely don't understand why some people are still bullish about LLMs

This is the correct approach. Also see: https://ezyang.github.io/ai-blindspots/

ukuinaon Using uv and PEP 723 for Self-Contained Python Scripts

Can uv install Python?

Is it possible to curl the uv binary and then invoke such a packaged script with --no-cache to run everything, including the Python installation, from /tmp?

ukuinaon Local-First and Ejectable

It is possible that life got in the way. Thymer did seem a great idea.

ukuinaon GPT 4.5 level for 1% of the price

They are still the benchmark against which all LLM progress and pricing is compared.

The general public will be using "ChatGPT" as a generic term for an AI chat interface for years.

ukuinaon Docs – Open source alternative to Notion or Outline

This looks neat! Will the self-hosted binary function in air-gapped environments?

ukuinaon An update on Mozilla's terms of use for Firefox

Horse Browser? https://gethorse.com

ukuinaon GPT-4.5

My site https://hackyournews.com does this!

Been keeping it alive and free for 18 months.

ukuinaon Tell HN: Y Combinator backing AI company to abuse factory workers

> If your management tracks your commits/velocity, look for a better job.

Don't most version-control systems support/automate this?

ukuinaon Material Theme has been pulled from VS Code's marketplace

> He thinks he owns literal hex codes

Pantone would like a word.

ukuinaon My LLM codegen workflow

> turns into legacy as fast as the agent can spit out new code.

This is an important point. Unconstrained code generation lets you witness accelerated codebase aging in real-time.

ukuinaon Show HN: Hackyournews.com v2

I've increased the font weight of the article title; this bolded look alongside additional vertical spacing really helps to add visual separation. Take a look and let me know how it can be improved further!

ukuinaon Show HN: Hackyournews.com v2

Would a thin line divider be preferable to whitespace?

ukuinaon Show HN: Hackyournews.com v2

Interesting; that story's already rolled off the front page, so I'm not seeing it currently.

ukuinaon Show HN: Hackyournews.com v2

I've added this, and it already looks much better. Thanks for the actionable suggestion!

ukuinaon Show HN: Hackyournews.com v2

I've increased the line height and font sizes as suggested in a sibling comment. Will investigate Mistral Small, thanks!

ukuinaon Show HN: Hackyournews.com v2

That's an interesting layout! Thanks for sharing, will investigate improving the whitespace utilization.

ukuinaon Show HN: Hackyournews.com v2

Yeah, trying to figure out how to get JS-heavy content without resorting to a headed browser.

ukuinaon We were wrong about GPUs

HashiCorp Nomad?

ukuinaon Show HN: Mikey – No bot meeting notetaker for Windows

Whisper accepts a system prompt.

ukuinaon Surrealist Compliment Generator

> You look like someone who has lunched poorly and who has no expectations for dinner.

Some juicy morsels in there!

ukuinaon DeepScaleR: Surpassing O1-Preview with a 1.5B Model by Scaling RL

I realize there is a team behind HN, but your sheer force of will keeps HN a shining example of discourse. Thank you for everything you do!

ukuinaon Google opens Gemini 2.0, its most powerful AI models to everyone

> The suite of models includes 2.0 Flash, which is billed as a “workhorse model, optimal for high-volume, high-frequency tasks at scale,” as well as 2.0 Pro Experimental for coding performance, and 2.0 Flash-Lite, which the company calls its “most cost-efficient model yet.”

> Gemini Flash costs developers 10 cents per million tokens for text, image and video inputs, while Flash-Lite, its more cost-effective version, costs 0.75 of a cent for the same.