Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bicepjai
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
19 ms
·
331.
▲
by
bicepjai
1y ago
Famous last words :)
332.
▲
by
bicepjai
1y ago
I would argue code is a meaningful sentence. So software writers is more appropriate :) ?
333.
▲
by
bicepjai
1y ago
World of 3d printing must be explored by everybody. I do see a world where we don’t have to order those plastic parts for almost anything. Creating them using 3d software and seeing them come to life gives the same feeling of launching a so
334.
▲
by
bicepjai
1y ago
What would you expect from AI model creating company CEO who figured their models were better with coding. These were the people who wanted alignment :)
335.
▲
We Tested Google Veo and Runway to Create This AI Film. It Was Wild [video]
(youtube.com)
16 points
by
bicepjai
1y ago
|
0 comments
336.
▲
by
bicepjai
1y ago
Great write up. The writing style reminds me of my new favorite phrase “imitation and innovation” the author has right amount of quotes and comparable new thoughts. This keeps the reader engaged and not feel like how much more we have to go
337.
▲
by
bicepjai
1y ago
Do you also have tokens per second metric ?
338.
▲
by
bicepjai
1y ago
Polars is amazing, but I would argue pandas is definitely not legacy :)
339.
▲
by
bicepjai
1y ago
They are saving time and tokens. They are rescanning the files locally using parsing and AST without sending tokens to the model which
340.
▲
by
bicepjai
1y ago
More focus on what to use rather when the whole file where the code snippet sits
341.
▲
by
bicepjai
1y ago
This is exactly why I read hacker news. Thanks for sharing.
342.
▲
Where Have All My Deep Male Friendships Gone?
(nytimes.com)
2 points
by
bicepjai
1y ago
|
1 comments
343.
▲
by
bicepjai
1y ago
I have lot of large gopro videos (have a 4 year old with all 4 years of gopro videos), that I painstakingly have to edit when my 256 GB card fills up (currently I have 500GB of vodes sitting) . No matter how careful you are while recording
344.
▲
by
bicepjai
1y ago
>>>CLASSIFIER-FREE GUIDANCE … During inference, you run once with a caption and once without, and blend the predictions (magnifying the difference between those two vectors). That makes sure the model is paying a lot of attention t
345.
▲
by
bicepjai
1y ago
My favorite tools are cline and roo. My experience says cline eats tokens like crazy and roo eats less. I don’t try Aider since I do like to watch the mesmerizing diffs (fire verification) :)
346.
▲
by
bicepjai
1y ago
After spending 1000s of dollars on 3 projects that I thought I could finish in a weeks :) with agents with cline, I reached similar conclusion. Best use cases I have seen it work is for good commit messages, templated documentation and curs
347.
▲
by
bicepjai
1y ago
Which agent is token hungry ? I notice cline is top on the list. Roo eats less than cline. Are there agents that we can configure how the interactions go ? How does Claude code compare to other agents ?
348.
▲
Such a Bad UI Pattern
(imgur.com)
2 points
by
bicepjai
1y ago
|
0 comments
349.
▲
by
bicepjai
1y ago
I just burned 200$ over a weekend trying to finish my pet project with cline and Claude. But soon I realized, this is not sustainable and I was half there. I tried Gemini pro 2.5 and with in 30 mins I hit 30$ and had to back off. Now trying
350.
▲
by
bicepjai
1y ago
Today I tried Gemini and have better experience with it on cline. So much rate limiting on Claude probably everybody using them. Also Gemini is expensive. I have already burned 200 dollars for one weekend incomplete project
351.
▲
by
bicepjai
1y ago
I completely agree-growing up, I admired the values I saw in American movies. It might sound amusing, but shaping young minds through film has been a key way the United States built its global influence. Hollywood movies have long served as
352.
▲
by
bicepjai
1y ago
I’m not necessarily suggesting that this content was written by AI, but the use of side headings for each point does make it seem AI-generated. That shouldn’t be an issue, though, since that’s exactly how I use AI myself.
353.
▲
by
bicepjai
1y ago
Did I just make it impossible for Anthropic to consider me for a job? :)
354.
▲
Anthropic AI Policy for Application
2 points
by
bicepjai
1y ago
|
2 comments
355.
▲
by
bicepjai
1y ago
Working on a fridge camera; no good solutions exist and I need it to avoid that feeling that I get when I dump my large bag of fresh vegetables.
356.
▲
by
bicepjai
1y ago
I love the take on fandom, this is how I would want it. While this article portrays fandom as a pure, innocent and positive force, my experience shows it can have a darker side. In places like South India, fandom often evolves: fandom becom
357.
▲
by
bicepjai
1y ago
Who ever says this have not done laundry :)
358.
▲
by
bicepjai
2y ago
Someone had to say this. All models are dump, but some are useful.
359.
▲
Ask HN: Is the US losing to China in tech creation and production?
8 points
by
bicepjai
2y ago
|
4 comments
360.
▲
by
bicepjai
2y ago
I have replaced all my search need with perplexity
More ›