vedmakk
vedmakk on Hacker News
> you have to let go of the idea that you first acquire skills and basic knowledge and then solve problems. The order reverses.
^ this
This is awesome! Thank you.
This is the kind of stuff I love about ancient architecture. It seems they were full of such clever things (or maybe only the few constructions which survived until today).
Its nice to see that some people still care about creating such thoughtful art for modern constructions. It seems that most building of our time are just optimized for fast and efficient construction.
I hope there are many more out there, so that Earth's Graham Hancock of the year 16000 has something to explore on his/her ayahuasca trip.
After reading that post it feels so basic to sit here, watching my single humble claude code agent go along with its work... confident, but brittle and so easily distracted.
That's less a URL shortener and more a URL dodgifier.
"Chrome wants to access 'secrets.txt'. Allow | Deny"
To everything in this article that states what AI cannot do... I would like to add a big fat "YET!" and remind everyone to buckle up...
Right now its convenient to look at Tailwind and discuss what their doing wrong.
But eventually most other business models will be stress tested in the same way - sooner or later.
missed a chance to use "gastimate"
This is really cool. I'm interested in the GenUI part. Is the web app itself static and the stories are generated on-demand?
Do you give gemini some UI components/templates to build with or is it just prompting to get consistent results across multiple stories?
If you want to go beyond a single one-off script, tou want to use it directly in your repo using the CLI tools or one of the IDE integrations (Copilot, Cursor, Zed, ...)
I agree on the voice mode... its really unusable now.
I feel like its been trained only tiktok content and youtube cooking or makeup podcasts in the sense that it tries to be super casual and easy-going to the point where its completely unable to give you actual information.
Same experience here. Shame.
I can definitely confirm this from my experience.
Gemini 3 feels even worse than GPT-4o right now. I dont understand the hype or why OpenAI would need a red alert because of it?
Both Opus 4.5 and GPT-5.2 are much more pleasant to use.
I don't get the Gemini 3 hype... yes it's their first usable model, but its not even close to what Opus 4.5 and GPT 5.2 can do.
Maybe on Benchmarks... but I'm forced to use Gemini at work everyday, while I use Opus 4.5 / GPT 5.2 privately every day... and Gemini is just lacking so much wit, creativity and multi-step problem solving skills compared to Opus.
Not to mention that Gemini CLI is a pain to use - after getting used to the smoothness of Claude
Code.
Am I alone with this?
So the US will be excluded from the SWIFT banking system? Heavy international sanctions will be put in place? Europe will send weapons and money to help Venezuela defend itself?
No? Oh... just checking.
Hey, interesting project!
Though human-in-the-loop is usually used in scenarios where control is held by said human (e.g. verification or approval).
The difference I'm curious about is agents being the primary caller, and humans becoming an explicit dependency in an autonomous loop rather than a human-in-the-loop system.
Posted by
vedmakk on Thursday January 01, 2026 @01:09PM
from the ask-hn dept.
Serious question.We're building agentic LLM systems that can plan, reason, and call tools via MCP. Today those tools are APIs. But many real-world tasks still require humans.So… why not expose humans as tools?Imagine TaskRabbit or Fiverr running MCP servers where an LLM agent can:- Call a human for judgment, creativity, or physical actions- Pass structured inputs- Receive structured outputs...
Reminds me of a quote from a few years back: "We are entering an era where we use AI to write blog posts from a few keywords for people who use AI to summarize a blog post into a few keywords".
If one would train an actual secret (e.g. a passphrase) into such a model, that a user would need to guess by asking the right questions. Could this secret be easily reverse engineered / inferred by having access to models weights - or would it be safe to assume that one could only get to the secret by asking the right questions?
"Verdichtung" is almost funny considering that Zurich is a city with not even half a million inhabitants. For example in the greater Moscow area live roughly 35 million people in an area about 1/4 the size of whole Switzerland with its 9 million people. THAT is "Verdichtung".
I was very happy to take my wife's name when we got married. A free choice I made. And I think there is nothing weird about it.
I changed my name to my wife's name when we got married. Where I live, everyone can choose if they want to keep their name or change it to either ones. So its a free choice.
AND: Hope gmail will rollout this feature asap, so I can FINALLY adjust my email address too.
My understanding is, that MCP Prompts and Slash Commands are "user triggered" whereas Skills (and MCP Tools) are "model triggered".
Other than that it appears MCP prompts end up as slash commands provided by an MCP Server (instead of client side command definitions).
But the actual knowledge that is encoded in skills/commands/mcp prompts is very similar.
Not enough long dashes across the article. Must be fake.
Yes, and the 1960s catalogs furniture seem still modern and would basically fit in a living room today. Timeless.
Wow, i randomly opened the 1960s catalog and the furniture looks just modern... they would perfectly fit in a todays living room.
Either time stood still in the furniture industry, they did timeless designs or the 60s trends are just coming back.
But I didnt expect that.
Because of the reasons presented in the actual article that was posted by op...