Skip to content

LLMs ‐ Yes and Why

Michael O'Brien edited this page Jan 28, 2025 · 4 revisions

In progress....

see https://obrienlabs.medium.com/running-reasoning-llms-like-the-deepseek-r1-70b-43g-locally-for-private-offline-air-gapped-259fa437da8f

LLMs are the future - embrace the change, use them to go to the next level. As in we dont write everything in machine code anymore. Or we don't throw our Sabo into the looms anymore. Or even - everyone has access to knowledge in books since the printing press replaced scribes.

Point 0: Enabling Introverted Programmers

Point 1: Purity - no more attitude-questions

  • We have all seen answers to a question we would like some expertise shared with us - but have received the following, like - "why are you doing this", "stop using x and use y"....

Point 2: Access to an expert that assists, summarizes, reviews your questions and code

Usage 1: Use as a partner

Usage 2: Limit use as a copy/paste developer

  • You still need broad base knowledge that needs daily rework/practice/use to avoid the depleting capacitor effect

Anecdotes

Analogies

Remember Bradley Cooper in "Burnt" - he was a Michelin level chef. When confronted with the Sous Vide style cooking (plastic bags in boiling water) - he initially rejected the idea. However, taking on a new method allowed him to explore areas previously uncharted.

In the same way that the computer has enabled us/me to run numerical simulations without being slowed down by paper. The GPU allows us to experience levels of mathematical insight that is 10-1000x accelerated.

Clone this wiki locally