There’s no exact French translation for “fair.” Words like justice or équité come close, but not quite. It’s an interesting reminder of how language reflects cultural perspectives and ways of thinking.
@mitchellh Mitchell, thank you for sharing. It’s interesting technique. Why do you manually write tests to verify it was right? Does it make sense to use AI for this?
@romaindewolff My tendency to treat ideas and systems with uncompromising rigor. In business people think I criticize them, but this brings the most actual value.
@tdinh_me Yes it's a big issue in France because I have to send all of my documents including Tax declarations to many random companies/individuals just to have services and chances to sign a contracts i.e. insurance, rent etc.
I tried Devin for development using their pay-as-you-go feature. While the agent itself seems promising, I was already charged $30 for just 2 PRs. At that cost, it’s hard to justify continuing with them.
Same story as with new programming languages: the first compiler is usually bootstrapped in something like C or C++, and once the language is mature enough, the next compiler is rewritten in the language itself.
Just watched the demo on https://t.co/8qeMB6s5Nw, turns out Zed is written in Zed itself. Astonishing and a bit magical. That’s what I love in programming.
RT to help Simon raise awareness of prompt injection attacks in LLMs.
Feels a bit like the wild west of early computing, with computer viruses (now = malicious prompts hiding in web data/tools), and not well developed defenses (antivirus, or a lot more developed kernel/user space security paradigm where e.g. an agent is given very specific action types instead of the ability to run arbitrary bash scripts).
Conflicted because I want to be an early adopter of LLM agents in my personal computing but the wild west of possibility is holding me back.