Hacker News: nisarg2

New comment by nisarg2 in "Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s"

nisarg2 — Sun, 04 Oct 2026 14:08:10 +0000

92% is halfway to 99%

Holds up pretty well

New comment by nisarg2 in "Claude Opus 4.7"

nisarg2 — Fri, 17 Apr 2026 03:21:08 +0000

Insane! Even Haiku doesn't make such mistakes.

New comment by nisarg2 in "We hid backdoors in ~40MB binaries and asked AI + Ghidra to find them"

nisarg2 — Sun, 22 Feb 2026 16:16:35 +0000

I wonder how model performance would change if the tooling included the ability to interact with the binary and validate the backdoor. Particularly for models that had a high rate of false positives, would they test their hypothesis?

New comment by nisarg2 in "Extending the context length to 1M tokens"

nisarg2 — Mon, 18 Nov 2024 18:14:14 +0000

At the end of the response they forget everything. They need to be fed the entire text for them to know anything about it the next time. That is not surpassing even feline intelligence.

New comment by nisarg2 in "The war on remote work has nothing to do with productivity"

nisarg2 — Mon, 23 Sep 2024 06:30:00 +0000

This is backwards. When covid started, the executives were saying that working from home will not affect productivity. They said zoom and slack are the new way to work. Some even said that productivity increased when people didn't have to commute. Some said people were working longer hours than earlier.

Now, all of a sudden, they realize their blunder. While the workers are happy, they are not. So now they will try to convince us that we are not productive enough, or that some of us are slacking, or that we have stopped innovating, or another 50 reasons for which going to the office will help us.

More recently, they have realized that they are not fooling anyone. So now they are threatening lower pay or layoffs for those who refuse to come back.

New comment by nisarg2 in "A website where you write a message to the next visitor"

nisarg2 — Sun, 23 Jun 2024 01:56:41 +0000

The harassment filter is a bit too sensitive. I mentioned something about some random "idiot" and it blocked me.

New comment by nisarg2 in "Trying out the panel-of-experts prompting strategy for LLMs"

nisarg2 — Tue, 21 May 2024 12:16:03 +0000

I have tried a similar idea before. I had two experts: one to propose a plan and another to critique it. It worked well, but often required 5 rounds of iteration to get to a great answer. The cost wasn't ideal and GPT4 still performed better on the first go, so I abandoned the idea.

I am curious how this approach compares to one-shot response from a larger model.