Susheel Daswani

Can't an open source project use AI to build out it's corpus to solve new problems (which will at some point be folded into a model via training)? I think the mistake here is conflating the purpose of open source. It's not to have folks visit docs for traffic, revenue or community - it's to provide a 'free as in speech, not as in beer' alternative to closed source software.

Thanks for sharing this! It's very easy for startup founders to simply default to 'speed at all costs', but this blog makes the case forcefully that i) pure speed can kill and ii) process and organization increases speed and, at the end of the day, customer happiness!

Congrats on the launch! How is this different than https://konghq.com/ ?

Cursor is an amazing product, but when writing Selenium scripts, several times it got into an 'infinite' loop of errors by adding something like the following to the script: firefox_options.add_argument("arg1") firefox_options.add_argument("arg2") firefox_options.add_argument("arg3") Until I told it to stop! 🤣

This seems like a great product! A few questions: 1) What does Depth offer that Github CoPilot + VSCode does not? 2) Can this be run 'on premise' / locally, i.e., integrated with a local model like Llama 3.3?