Just hit 10K subscribers on YouTube! 🎉 . My early videos teach best practices for building LLM agents. Then I realized: why not automate making the videos themselves?. So my latest videos are made by an LLM agent I built to teach various technical topics. BEST testimony for my

Join millions who have switched to Grok.

RT @Hesamation: A banger on how to publish the ultimate journal approved research paper in 2025:

RT @zoink: tell me again about how locked in you are

RT @quantbeckman: This is the reason why Renaissance hires PhDs to clean data 😬

RT @DimitrisPapail: Why is cross-entropy a good loss for language pretraining?. caveat: this is all known btw; interestingly, even though t….

Claude Code is working for me 24/7, even when I sleep. Here is the command I just sent to it. I'm doing some grunt work of data processing. It has already processed a few to my satisfaction. So I just ask it to keep grinding while I'm going to sleep.

RT @Guangxuan_Xiao: I've written the full story of Attention Sinks — a technical deep-dive into how the mechanism was developed and how our….

I've switched to Claude Code from Cursor, but I bet the Gemini CLI will catch up. Here's my take:. For coding agents, two things matter most: (a) Solid agent design (b) Generous token usage for a top-tier LLM.Everything else (codebase indexing, specialized models that applies.

RT @joecarlsonshow: This is why I don’t trim Microsoft.

RT @random_walker: Many people have told us that this is one of the most memorable opening paragraphs they've come across in a book. 🙏 The….

RT @EverythingOOC:

RT @techdroider: Google Cooked 🤫

RT @Kangwook_Lee: @DimitrisPapail (financial) risks.

RT @DimitrisPapail: We don’t even understand a two layer perceptron.

Just tried out Grok 4 after paying $30. Overall disappointed—coding's not bad but slightly worse than Gemini 2.5 and Claude 4. Doc writing feels terse and keeps running into weird access issues.

RT @mayfer: why are they like this

RT @GithubProjects: $ git push origin master --force

RT @SimonXinDong: It turns out,. > GRPO is performing the arithmetic mean --> token-level scaling.> GSPO is performing the geometric mean -….

The video I posted 2 weeks ago: