v3ss0n

Joined 12 years ago

Karma

1,580

Submissions

1,014

Member since

Aug 7

https://hexcode.tech We are a team of professional opensource enthusiast from Yangon, Myanmar. We love developing high quality software and we have 20 Years of experience developing various technologies.

View on Hacker News

Recent submissions

Comment · 2 days ago

And -- that is our reality.

Comment · 3 days ago

on average it is fine. On difficult tasks it still does hallucinate. - like decision based on huge policy documents ( context reaching around 900k)

A lot less these days tho.

Comment · 3 days ago

Yo Dawg , you love hallucination so we develop drugs with LLM.

Comment · 4 days ago

Onlything is a lot of cextensions are super slow there. that is only bottleneck left which could be solved with resources that those guys spent on experimental jit.

Comment · 4 days ago

They should had just port everything to pyPy.

Comment · 4 days ago

Git Worktress, Sub Agent WIth Team Leader agents. I get things done alot faster, You are just doing it wrong.

Comment · 5 days ago

https://news.ycombinator.com/threads?id=v3ss0n#:~:text=alrea...

Sorry i reply to wrong comment.

Comment · 5 days ago

already , i self hosted Deekseek 4.1 Flash , Qwen 3.8 Flash Next , Glm 5.3 flash , tested it , all benchmaxxed to the point that all of them can draw that well , much better than other logos .

Comment · 5 days ago

Sorry I didn't knew it was started here and , lol I didn't know it was viralized by Simon of django! Now these days , many LLMs are benchmaxxed to it.

Comment · 10 days ago

I hated react and don't use at all but credit to credit due

Comment · 11 days ago

PyTorch .. React ..

That two libraries already build Frontend and AI.

Comment · 14 days ago

Me too Deekseek harness , is sleek , solid , and flexible.

Comment · 16 days ago

Try pangolin for tunneing , it works so well.

Comment · 17 days ago

FBI!! OPEN UP!