Viewing profile — 1899-12-30

It's much more censored - in some aspects - than k2.7.

you can get a rough version via artificialanalysis's cost per task https://artificialanalysis.ai/?cost=intelligence-vs-cost-per...

makes it easy to discard someones opinion when they hit you with 'geniunely'[0] [0]: https://trends.google.com/explore?q=genuinely&date=all&geo=U...

The days of capital light companies might be over for the near future.

*Minnesota becomes first state to ban the use of prediction markets as a loophole for sports gambling.

Costco buisness centers are usually a lot quieter, and don't have the free sample venders clogging up the aisle. Memberships work for both.

It's cheaper to pay that than rewrite in a better language.

The key to non-disruptive LLM integration is using it in a purely additive way, supplementing a feature with functionality that couldn't be done before rather than replacing an exi…

There is, it's called a narrow bank and it would probably pay more interest than normal banks. Unfortunately they have been mostly outlawed thanks to lobbying by the banking indust…

M$ already has a program like this doesn't it? Microsoft™ 365™ Copilot™ agents™ Weird amounts of overlap between the two.

have you tried the emacs package agent-shell?

I pray for whoever has to review the slop I've generated.

https://eqbench.com/spiral-bench.html

A crash in the stock market doesn't necessarily mean a crash in the real market, The AI bubble burst being dot com style vs a gfc debacle depends on how much critical financial inf…

If you talk to someone you know, they'll hold it against you for the rest of your life. If you talk to an LLM(ideally locally hosted) the information dies with the conversation con…

Or subagents that check each link to see if they verify the actual claims the links are sourced for.

It's a massive red flag if the company is using it to inflate their revenue, or using debt based tools. Stock swap based vendor financing is a bit less red since nvidia/amd could g…

That's a difference in the system prompt, not the model itself.

iirc all you have to do is ask them if they think kim jong un looks fat

LLMs behaviour is a lot closer to primary orality than literacy.

I fear for the day that bot responses are not identifiable by em dashes and "not x but y" structures.

Delphi sucks, I don't recommend anybody use it when you could just use c#.

Chatbots would be great for discoverability, especially if you have a ton of docs that nobody seems to read.

Are there any MCP servers that enrich the context with data from a language server and fetches relevant documentation based on the types provided from that information?