Michael Erasmus

@ArmanHezarkhani MLE

If you get through all of this material, you'll be an ML Engineer (MLE) I'm thinking of leading a study group through this content If you're interested in joining, comment "MLE" below

@georgesttock nano

Nano banana + MakeUGC + Veo3 = Ad Factory This agent creates 100s of ads every day - No actors - No product - No $10k/month agencies You're able to clone your comepitors ads with AI Paste there ad -> Pick an avatar and regenerate. Comment "NANO" and I'll send you the agent + the full playbook (must be following)

23948h of debugging later. Kill me.

Last week, in our study group "AI from Scratch", we finished reading "Build a Large Language Model" by Sebastian Raschka. We voted on what to read next, and "AI Engineering" by @chipro won by a LOT! 🔥 Here are the details: 📅 First Meeting: Sunday, Feb 16th at 2:00 UTC 📖 How it works: During the week, we each read and study a specific chapter independently—this time, Chapter 1. On Sunday, we meet for a 90-minute call where someone presents the chapter, walks through the code, and we all discuss. 🔗 How to join: Join our Discord channel with 1,000+ learners! discord.gg/ZmM2V4k2qF?eve…

@behrouz_ali @mirrokni @threadreaderapp unroll

Attention has been the key component for most advances in LLMs, but it can’t scale to long context. Does this mean we need to find an alternative? Presenting Titans: a new architecture with attention and a meta in-context memory that learns how to memorize at test time. Titans are more effective than Transformers and modern linear RNNs, and can effectively scale to larger than 2M context window, with better performance than ultra-large models (e.g., GPT4, Llama3-80B).

We're entering the golden age of learning. It's still early but I can imagine with these kinds of personal AI tutors the speed at which you can learn new things will just keep increasing.

@OfficialLoganK @pydantic @OpenAI @colinhacks @OfficialLoganK How do you specify your project/location using the OpenAI SDK?

Gemini in the OpenAI SDK: we now support Structured Output requests through the OpenAI SDK to Gemini models, including with support for @pydantic & Zod! 🔀

Coding interviews should be using ai. If you disagree you literally don’t understand what ai is for Sure you can use it to fill gaps & find easy answers More importantly you can use it to accelerate and do much bigger things Failing to select for this skill will ruin tech co’s

This cuts deep...

$ hawk tuah

Conversations about unconventional things to do are a fun way to test an LLM, because the more you dig the more you're inadvertently encouraging it to hallucinate based on some weird and interesting context it's already added to the conversation

If you haven't decided yet who to vote for, here's why I think you should vote for Harris.

@santiviquez @rasbt I'm interested!

I’m thinking about organizing a study group to go @rasbt's latest book, Build a Large Language Model from Scratch. The idea is to read a chapter beforehand and then meet each week on Zoom to discuss, compare notes, and learn from each other. I think we can cover it in 6 weeks!

I think often about Wolfram’s conclusion nearly two years ago that ChatGP shows “that human language (and the patterns of thinking behind it) are somehow simpler and more “law like” in their structure than we thought. ChatGPT has implicitly discovered it.” Seems worth considering

This is awesome

Are you kidding me? Look at this:

@t3dotgg I switched to Linux a couple of years ago. Same thing happened to me and I lost like 3 days trying to recover. Switched back to a MacBook and haven't looked back since.

Tried to wake from sleep. Sign in screen wouldn’t accept inputs. Tried rebooting. It will no longer boot. The Linux experience 🫡

a machine learning engineer should be able to debug a pydantic validation error, debug an OpenAI completions error, debug model drift, debug Airflow, debug a FastAPI method, debug a Spark dataframe, debug a put to a feature store, specialization is for insects - Robert Heinlein

Me installing all the python packages I need before starting coding on a new project...

Yep, but then you're not even counting college. I always tell people think hard about having kids, it's the most expensive decision you'll make.