MikhailTal

6 ms·

1.
by MikhailTal 9d ago
Aware that this is essentially a soon to be deprecated service. As any vendor you need to know the life expectancy of it, plus forecast/predict any pricing changes. You really do not want to become dependent. Even if competitors exist,
2.
by MikhailTal 9d ago
Again, this is HN bubble stuff. The vast majority of people do not mind oauth or account.
3.
by MikhailTal 10d ago
> if you hook up to the fediverse No this is the hard problem. HN people live in a bubble. Anything else than signup with oauth and then having access to everyone and its too hard and 99% of people will click away.
4.
by MikhailTal 21d ago
As usual, the real tldr is in the privacy policy https://generalresearch.com/supplier/privacy-policy/notice-t... > General Research Laboratories, LLC (“GRL”) is an aggregator of market research surveys in multi
5.
by MikhailTal 23d ago
Very fascinating, super interesting engineering. Although i do find it very funny how they just bypass a massive vulnerability, basically zero data isolation (even between good actors, let alone bad ones) with 3 sentences. Only in the llm s
6.
by MikhailTal 25d ago
Not all agents have access to a sandbox/cli/code execution environment to run arbitrary api calls etc. MCP helps by essentially having another tool call without needing a sandbox. If you do have a sandbox, then might as well do co
7.
by MikhailTal 1mo ago
What is the difference with Livekit Gateway? https://livekit.com/blog/introducing-livekit-inference Or even something more managed like Vapi?
8.
by MikhailTal 1mo ago
This is terrible. Models have been RLed on looking at the previous tool call chain, and reasoning. No chance this does not reduce performance. The point of compaction is that it also includes useful signal from the tool outputs itself so a
9.
by MikhailTal 1mo ago
When a company gives away service a heavily subsidized service as a promo, the full cost of serving it (compute) can get classified as sales and marketing instead of just cost of revenue, which makes your gross margin look better!
10.
by MikhailTal 1mo ago
https://github.com/e2b-dev/e2b
11.
by MikhailTal 1mo ago
> Google DeepMind: We are building strong momentum: Flash is in high demand, our Cyber model is live, and Gemma models have surpassed 900M+ downloads Considering these are the best stats they could find, gemini usage+general situation mu
12.
by MikhailTal 2mo ago
It's a smart approach, definitely interesting. It all hinges on quality of course which im not convinced. τ³-Banking is the only one which you show better accuracy and cheaper. If i'm reading the blog results right, for deepswe an
13.
by MikhailTal 2mo ago
this is /goal in claude code/codex. also basically a slightly improved ralph loop
14.
by MikhailTal 2mo ago
is this really that surprising? Exploitgym prompts are tuned for a model to do everything it can to achieve a cybersec/exploit task. And we know that models are good at finding vulverabiltiies. Its just random that the sandbox itself w
15.
by MikhailTal 2mo ago
the OP asked why, you more described the how. What is the benefit of all this?
16.
by MikhailTal 2mo ago
"Lottery winner says you should not gamble"
17.
by MikhailTal 2mo ago
Also, the OP just does not understand how the market works anyway. Surely if it was obvious that investing in fresh IPOs is a bad move, all of the big boys (banks, hedge funds etc) would short them to the point of equalising anyway. Maybe n
18.
by MikhailTal 3mo ago
This is not a new situation. This was happening also when good vision models like alexa net were coming through, especially for OCR. Companies had choice between cloud or self hosting with GPUs. But turns out, problem is usage patterns. You