monte

monte

@montepy.in

Progress over perfeciton Building: https://saral.club Other project: https://longpost.in Blog: https://montepy.in

128 followers1.1K following493 posts

monte
@montepy.in

Now that claude code is taking over my coding, I finally have time to handle office politics

monte
@montepy.in

Up until now, I used to write code for myself, but now I have to write in a way which makes it easy for AI.

monte
@montepy.in

The fact that all the AI companies are coming with minor version upgrades like chatgpt5.2, claude4.6 indicates that model progress has plateaued. But more innovation is on the engineering side. Claudebot aka moltbot aka openclaw is an example of that. (1/2)

monte
@montepy.in

Fundamentally, openclaw didn't implement any new ideas. But what makes it good is integrations. Setting it up with the things that I use regularly like whatsapp is what makes it good. (1/2)

monte
@montepy.in

Today I learned, that in python `yield from`, although similar, is very different from: [image_1] (1/2)

monte
@montepy.in

Fans of black mirror will realise that new ads of anthropic are much like the common people episode

monte
@montepy.in

I believe tools like code rabbit are much better than cursor. Not for productivity, but for junior software developers. You are still thinking and writing code yourself, but you have reviewer who gives you a valuable feedback. #BuildInPublic

monte
@montepy.in

When you pass a string to Date constructor, it doesn't raise an error. It doesn't return NaN. In fact, it doesn't even return the same string, but instead capitalise each word. Just #javascript #BuildInPublic [image_1] (1/2)

monte
@montepy.in

Filling out stackoverflow developer survey felt like paying homage to a fallen hero. It might possibly the last time we get to fill it. #BuildInPublic

monte
@montepy.in

Everybody who has used chatgpt feels this and now there is evidence also to support. It's making you dumber. Time to go back to neovim. www.media.mit.edu/publications... #BuildInPublic

media.mit.eduYour Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task – MIT Media Lab This study explores the neural and behavioral consequences of LLM-assisted essay writing. Participants were divided into three groups: LLM, Search Engine…
monte
@montepy.in

I know i am late to the criticism of liquidass, but it still boggles my mind that a trillion dollar company which is known for it's design shipped this. #BuildInPublic

monte
@montepy.in

A lot of new programmers are going to grow up without the feeling of "discovering a better solution" #BuildInPublic

monte
@montepy.in

The release of liquid glass reminds me of the famous quote from the dark knight, "you either die a hero or live long enough to see yourself copy Microsoft" #BuildInPublic

monte
@montepy.in

Writing tests is admitting your code can have bugs. Don't doubt yourself and go ahead with the production deployment. Follow me for more programming tips. #BuildInPublic

monte
@montepy.in

TIL, that you should treat emails as case insensitive, don't expect users to always type in correct emails #BuildInPublic

monte
@montepy.in

The greatest feature of apple intelligence is the option to turn it off. Had to turn it off because it was freezing wezterm #BuildInPublic

monte
@montepy.in

Today I learned, you can't use the relative paths (or full path) for image in open graph tags. It has to be a full url with https. #BuildInPublic

monte
@montepy.in

If someone can recommend resources on fine tuning llm for local (personal use). It would be helpful. #pythonprogramming

monte
@montepy.in

AI infra space is so crowded right now. Every week someone is coming up with something and I can't figure out if it's hype or genuine.