gandalfgeek

5 ms·

31.
by gandalfgeek 1y ago
The title of the study is provocatively framed and the actual findings don't live up to it. I made a short video explaining it-- https://www.youtube.com/watch?v=hLDCi0VwyiQ
32.
Natlang Code(vivekhaldar.com)
2 points by gandalfgeek 1y ago|0 comments
33.
The Mac-vs-PC Story Playing Out in CLI Agents(vivekhaldar.com)
2 points by gandalfgeek 1y ago|0 comments
34.
Plato's Cave of AI Agent Evals(vivekhaldar.com)
1 points by gandalfgeek 1y ago|0 comments
35.
Fixing Auth for Personal AI Agents(vivekhaldar.com)
1 points by gandalfgeek 1y ago|1 comments
36.
The programmer's climb(vivekhaldar.com)
8 points by gandalfgeek 1y ago|0 comments
37.
by gandalfgeek 2y ago
Very cool. Even cooler to see it upload itself!!
38.
by gandalfgeek 2y ago
There is no fundamental blocker to agents doing all those things. Mostly a matter of constructing the right tools and grounding, which can be fair amount of up-front work. Arming LLMs with the right tools and documentation got us this far.
39.
Dijkstra would've (incorrectly) HATED vibe-coding(vivekhaldar.com)
3 points by gandalfgeek 2y ago|1 comments
40.
"But LLMs are not deterministic "(vivekhaldar.com)
3 points by gandalfgeek 2y ago|7 comments
41.
The case against local LLMs(vivekhaldar.com)
2 points by gandalfgeek 2y ago|0 comments
42.
by gandalfgeek 2y ago
Wet streets cause rain.
43.
Human evals are outsourced vibe checks(vivekhaldar.com)
1 points by gandalfgeek 2y ago|0 comments
44.
Spec-Driven Vibe-Coding(vivekhaldar.com)
3 points by gandalfgeek 2y ago|1 comments
45.
AI resistance considered harmful (for programmers)(vivekhaldar.com)
1 points by gandalfgeek 2y ago|2 comments
46.
by gandalfgeek 2y ago
Most charities on that list are on sharp sides of deeply polarizing culture war issues and it is not at all clear that their causes align with the “American Dream”. If the last election was any indication then more than half the country exp
47.
by gandalfgeek 2y ago
I don't understand why he has such an axe to grind. Is there some historical baggage here? Of course there are plenty of problems with the current state of AI and LLMs, but to have such a preconceived pessimistic outlook that can'
48.
by gandalfgeek 2y ago
Love Fogus, but zero mention of how LLMs impacted programming in 2024?
49.
I use AI (late 2024)(vivekhaldar.com)
2 points by gandalfgeek 2y ago|0 comments
50.
by gandalfgeek 2y ago
This is a BS story. Pretty much every public company, at least every bigtech company, follows the same conventions -- don't say incriminating things in chat, trainings for "communicate with care" (definitely don't say &q
51.
by gandalfgeek 2y ago
There was one slogan that was repeated during COVID that perfectly encapsulates the degeneration and capture of science: "Follow the science". That's not how science works. Religions are "followed". Science is based
52.
by gandalfgeek 2y ago
Thanks for the pointer! "Call by meaning" sounds exactly like LLMs with tool-calling. The LLM is the component that has "common-sense understanding" of which tool to invoke when, based purely on natural language understa
53.
by gandalfgeek 2y ago
(former Googler) It was really special to see how this pair basically laid out the foundations of large-scale distributed computing. Protobufs, huge parts of the search stack, GFS, MapReduce, BigTable... the list goes on. They are the only
54.
Subject-object-verb, or– how we use AI(vivekhaldar.com)
5 points by gandalfgeek 2y ago|0 comments
55.
by gandalfgeek 2y ago
Keep seeing DJI drones at local police dept open houses. They even have a "drone unit" that specializes in SAR, hazardous recon type scenarios. Given extensive existing use throughout US local law enforcement, fire depts etc, not
56.
by gandalfgeek 2y ago
Agree that your use-case is different. The papers above are dealing mostly with adding a domain-specific textual corpus, still answering questions in prose. "Teaching" the LLM an entirely new language (like a DSL) might actually
57.
by gandalfgeek 2y ago
This was kind of conventional wisdom ("fine tune only when absolutely necessary for your domain", "fine-tuning hurts factuality"), but some recent research (some of which they cite) has actually quantitatively shown that
58.
by gandalfgeek 3y ago
#1 motivation for RAG: you want to use the LLM to provide answers about a specific domain. You want to not depend on the LLM's "world knowledge" (what was in its training data), either because your domain knowledge is in a pr
59.
by gandalfgeek 3y ago
> No HBM because they use tons of fast SRAM instead. Isn't that the main driver for performance here? No doubt fast SRAM helps, but from a computation pov imho its that they've statically planned computation and eliminated all
60.
by gandalfgeek 3y ago
They're calling the lie on needing bleeding edge hardware for performance. 5 yr old silicon (14 nm!!) and no hbm. Their secret sauce seems to be an ahead-of-time compiler that statically lays out entire computation, enabling zero conte

More ›