Mahesh Sathiamoorthy (@madiator) / X

Mahesh Sathiamoorthy

4,592 posts

Mahesh Sathiamoorthy

@madiator

RL Environment Curation. Data Curation (OpenThoughts). Post-training. CEO @bespokelabsai. Ex-GoogleDeepMind.

Inside a RL Environment

Joined February 2008

Pinned
Mahesh Sathiamoorthy
@madiator
Jan 28, 2025
We are announcing Open Thoughts, our large-scale open-source effort to curate the best open reasoning datasets! DeepSeek-R1 is amazing but we still don't have access to high-quality open reasoning datasets. These datasets are crucial if you want to build your reasoning models!
234K
Mahesh Sathiamoorthy
@madiator
Jul 25, 2023
This makes me sad: observablehq.com/@ayhanfuat/the…
4.5M
Mahesh Sathiamoorthy
@madiator
Sep 25, 2025
~2012 must have been like a crazy time. Neural networks were considered nonsense by most people. Hinton, LeCun and Bengio were the amongst the very few people who kept it all alive for decades. Alex Krizhevsky, a mad coding genius and a socially aloof kid, shows up to Hinton's
485K
Mahesh Sathiamoorthy
@madiator
Apr 27, 2023
Someone wrote article on Transformers from scratch: e2eml.school/transformers.h… I mean, really from scratch..
193K
Mahesh Sathiamoorthy
@madiator
Sep 20, 2025
I love how he has articulated what he wants clearly. But people still don't seem to understand what this is. It's not RAG, it's not notebooklm, it's not lmstudio. It's not a solved problem at all, contrary to what anyone claims. I have also always wanted to something like this,
429K
Mahesh Sathiamoorthy
@madiator
Dec 27, 2024
I think the community is excited about DeepSeek v3 not because it's yet another powerful model but because it's a story of human ingenuity in the face of constraints. Despite all the restrictions due to export control and limited budget, the humans of DeepSeek have created a
247K
Mahesh Sathiamoorthy
@madiator
Feb 9, 2025
Some people asked me about a resource for learning about Transformers. Here's a good one I am sharing again -- it covers just about everything you need to know. brandonrohrer.com/transformers Amazing stuff. It's totally worth your weekend.
112K
Mahesh Sathiamoorthy
@madiator
Sep 17, 2025
> be Google > understand how important machine learning and deep learning is and create Google Brain in 2011 when everyone in the industry is screaming "big data". > acquire hinton's company called DNNresearch in 2013, which is basically Alex Krizhevsky and Ilya Sutskever and
Zephyr
@zephyr_z9
Sep 13, 2025
> be Google in 2017 > small team drops “Attention Is All You Need” on arXiv > execs nod politely, go back to selling ads for socks > let Transformer gather dust for 5 yrs like a vintage Beanie Baby > be Noam Shazeer, OG wizard > quits, builds AI-boyfriend app
195K
Mahesh Sathiamoorthy
@madiator
Dec 10, 2023
Update: I recently left my dream job at Google DeepMind to start something new. It was not an easy decision, given how amazing Google DeepMind has been and how much fun it has been to work with the incredible set of people there. I want to thank all my colleagues, managers, and
295K
Mahesh Sathiamoorthy
@madiator
Apr 1, 2023
If you are a PhD student, you should check out the book called "How to take smart notes". I wrote a bit about this book and what I have learned about note-taking in newsletter.smarter.blog/p/how-to-take-….
209K
Mahesh Sathiamoorthy
@madiator
Oct 15, 2025
DeepMind was acquired by Google for north of 500M back in 2014. Zuckerberg/Meta had also apparently bid for DeepMind, and apparently at a much higher number, but still lost because Demis and co wanted to go to Google. Maybe that explains the lengths he is going to get AI talent
128K
Mahesh Sathiamoorthy
@madiator
Jan 15, 2025
We are happy to announce Curator, an open-source library designed to streamline synthetic data generation! High-quality synthetic data generation is essential in training and evaluating LLMs/agents/RAG pipelines these days, but tooling around this is still entirely lacking! So
GIF
201K
Mahesh Sathiamoorthy
@madiator
Dec 24, 2024
People are panicking with O3 doing really well on ARC-AGI, so wanted to share some perspectives. 1. OpenAI trained on 75% of the training set of ARC-AGI. On X there has been a lot of discussion around this, but wanted to share what i think. So training on train set is fine by ML
OpenAI's o3 Sets New Record, Scoring 87.5% on ARC-AGI Benchmark
From maginative.com
215K
Mahesh Sathiamoorthy
@madiator
Sep 19, 2023
Wow, run LLMs like BitTorrents! petals.dev github.com/bigscience-wor…
110K