HN user
Could the (not so perfect but technically simple) solution be to transform the style of content under each tag to the correct expected style for the tag, via a smaller or purpose-built LLM, before the data stream is fed into the main LLM? Perhaps the two LLMs can be co-trained to keep the overall quality of the output stable while role confusion is minimized.
exactly
Will there be refund?
You sound bitter. May be you don't realize that.
To be fair, the author does mention the huge difference between Gemma 3 and Gemma 4 on Tau function calling benchmark.
Very interesting. How does this approach work for complex agentic workflows where the LLM is expected to orchestrate across multiple tools (such as when using MCP)? Or is this mainly for simple cases like the ones presented in the blog post?
Dwarkesh*
The current CEO is Sridhar Ramaswamy
Interesting. Is there a source for this claim?
What does this mean for those holding CA municipal bonds and bond funds?
chessprogramming.org is a treasure trove of knowledge on building chess engines. With its help, I wrote one in C++ a few years ago that got quite good (2100+ rating on FICS but that's nowhere close to the likes of Stockfish). In fact, writing a reasonably strong chess engine is straightforward (and incredibly fun) but at the top end of strength, there's immense depth, and after a point making improvements gets increasingly resource intensive (tuning params, running experiments to verify strength gain all takes a lot of compute).
Chess programming is also extremely addictive. On forums like talkchess.com, you see folks hanging out who have been doing it for decades (most of them are also super helpful to newbies).
Just completed part 1 (used Python instead of Java to spice it up a bit) and it was a great experience! Looking forward to working through part 2, but it will be a while before I can get around to it. I hope more books follow this engaging writing pattern (when it makes sense). Kudos to Bob Nystrom for this great book.
FWIW, SpaceX uses C++ for flight control software. It actually makes more sense for such hard realtime applications.
And yet there are hardly any software written in Haskell or Ocaml that are widely used or have any notable positive impact on the modern digital world, compared to those written in languages like Go, C++ or Python.
edit: ps: Big fan of OCaml, but have since moved on to Go and Python for getting things done in the real world.
The sci-fi novel "Project Hail Mary" by Andy Weir. Captivating, and a lot of fun.
You are joking right?
Curious about this scenario:
Bob works for company A which is a vendor for companies B, C and D. Most of company A's revenue stream is dependent on how well B, C and D are doing. Bob has insider knowledge of Company A's finances using which he trades stocks of B, C and D.
Is this considered insider trading or not?
But Trump [1] also says the non-white congresswomen known as "the squad" should go back to where they came from. Even riled up his supporters [2] to chant "send her back".
You know "go back to where you came from" is a common racial insult hurled at non-white immigrants decades or generations after they naturalize in the USA?
[1] https://www.youtube.com/watch?v=bMZnemqsCdI [2] https://www.youtube.com/watch?v=CqLxxbjVG_k
Tell me this: When Trump riled up his supporters to chant "send them back" when referring to the American congresswomen known as "the squad", was it racist or not?
"Go back to where you came from" is a common racial insult non-white immigrants endure even decades or generations after they naturalize.
Google sent me to: https://www.scientificamerican.com/article/the-case-against-...
"The most devastating argument against the Copernican universe was the star size problem. When we look at a star in the sky, it appears to have a small, fixed width. Knowing this width and the distance to the star, simple geometry reveals how big the star is (right). In geocentric models of the universe, the stars lie just beyond the planets, implying that star sizes are comparable to that of the sun (below). But Copernicus's heliocentric theory demands that the stars be extremely far away. This in turn implies that they should be absurdly large—hundreds of times bigger than the sun (bottom). Copernicans could not explain away the anomalous data without appeals to divine intervention. In reality, the stars are far away, but their apparent width is an illusion, an artifact of the way light behaves as it enters a pupil or telescope—behavior that scientists would not understand for another 200 years."
nimgrep's alright. I was thinking of code search on, say, github. I feel the additional complexity was unnecessary in an otherwise well designed language.
One thing I find annoying about Nim is the case insensitivity [0]. There was no strong reason for this "feature" and literally no mainstream language does it. Moreover it makes code search a pain.
[0] https://nim-lang.org/docs/manual.html#lexical-analysis-ident...
Some junior engineers do (most don't). There was this thread from a while ago [0] where someone negotiated a 300k offer. I believe they were a new/recent grad.
Most of the time it's due to having some leverage (multiple FAANG+ offers).
This is very common in SV. Many junior/mid-level engineers at FAANG companies beat this figure (when accounting base + stock + bonus).
Did you confuse HN for Breitbart?
The article states definitively that Mars does not have tectonic plates. This is news to me because hitherto I thought there is no evidence either way.
H1B is a "dual intent" visa.
You're attacking a strawman. That quote makes sense in the context (basically the whole previous paragraph):
The rocket as it performs ascent and then descent undergoes some serious accelerations (way higher than a typical plane), while being built extremely delicately to minimize the mass (much tighter margins than an airliner). The rocket engine undergoes much higher stress than a jet engine. Hence all that stress can cause material fatigue and small malfunctions, while the margin for error on a rocket is extremely small. Hence if e.g. it turns out that the risk of mission failure on these reused boosters increases substantially after every use, all the anticipated savings from reusability may be quickly wiped out by one or two lost missions. And it should be noted, that the cargo that goes into space is often MUCH more expensive than the rocket itself.