Python (3 blogmarks)

There is a big grain of salt to be taken with any ranking like this. Nevertheless I find it interesting to 1) see the quick jump for Python in recent years that correspond with LLM proliferation and 2) the sharp decline over the past year.

The rise makes sense if you consider that a lot of AI and LLM tooling (e.g. pytorch) and general data science and data processing tooling are in the Python ecosystem. I'm not sure what the decline means. If I had to make a guess — LLM models and harnesses have gotten so good in the past year that developers are doing a lot less google searches and a lot more prompting as they use Python (and other languages).

Ruby -- the language that my career is mostly built on -- is down in the 30th spot.

Andrej Karpathy’s latest blog post (he doesn’t post very often, so you wanna check it out when he does) presents an implementation of a GPT in 200 lines of python with no dependencies.

This file contains the full algorithmic content of what is needed: dataset of documents, tokenizer, autograd engine, a GPT-2-like neural network architecture, the Adam optimizer, training loop, and inference loop. Everything else is just efficiency.

These are three books recommended to read, in this order (but, of course, choose your own adventure):

• Effective Python
• Python Cookbook
• Fluent Python