Monte Thakkar (@montethakkar) 's Twitter Profile
Monte Thakkar

@montethakkar

Building @joinrosebud ๐ŸŒน | ex @curiotools @nftsalesbot @expedia @pillow

ID: 1630740098

linkhttps://montethakkar.com/ calendar_today29-07-2013 16:30:25

8,8K Tweet

1,1K Followers

1,1K Following

Chrys Bader (@chrysb) 's Twitter Profile Photo

๐ŸŒน ๐—ช๐—ฒ ๐—ฟ๐—ฎ๐—ถ๐˜€๐—ฒ๐—ฑ $๐Ÿฒ๐—  ๐˜๐—ผ ๐˜€๐—ฐ๐—ฎ๐—น๐—ฒ ๐˜๐—ต๐—ฒ ๐˜„๐—ผ๐—ฟ๐—น๐—ฑโ€™๐˜€ ๐—น๐—ฒ๐—ฎ๐—ฑ๐—ถ๐—ป๐—ด ๐—”๐—œ ๐—ท๐—ผ๐˜‚๐—ฟ๐—ป๐—ฎ๐—น Five years ago, Sean Dadashi and I realized the attention economy was doing more harm than goodโ€”keeping kids and adults glued to screens, chasing dopamine. As technologists, we

๐ŸŒน ๐—ช๐—ฒ ๐—ฟ๐—ฎ๐—ถ๐˜€๐—ฒ๐—ฑ $๐Ÿฒ๐—  ๐˜๐—ผ ๐˜€๐—ฐ๐—ฎ๐—น๐—ฒ ๐˜๐—ต๐—ฒ ๐˜„๐—ผ๐—ฟ๐—น๐—ฑโ€™๐˜€ ๐—น๐—ฒ๐—ฎ๐—ฑ๐—ถ๐—ป๐—ด ๐—”๐—œ ๐—ท๐—ผ๐˜‚๐—ฟ๐—ป๐—ฎ๐—น

Five years ago, <a href="/seandadashi/">Sean Dadashi</a> and I realized the attention economy was doing more harm than goodโ€”keeping kids and adults glued to screens, chasing dopamine. As technologists, we
Monte Thakkar (@montethakkar) 's Twitter Profile Photo

"Combining human craftsmanship with AI tools to build better software." This is one of the best ways I've heard software engineering be described in the age of AI zed.dev/agentic-engineโ€ฆ

Monte Thakkar (@montethakkar) 's Twitter Profile Photo

Proud to be on the team that's on the cutting-edge of using AI for good. Love this initiative and excited to attract like minded folks that truly care about this stuff. This work is very important to ensure we use AI to process emotions and be a thought-partner in all the ways

Andrej Karpathy (@karpathy) 's Twitter Profile Photo

Excited to release new repo: nanochat! (it's among the most unhinged I've written). Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single,

Excited to release new repo: nanochat!
(it's among the most unhinged I've written).

Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single,
Brie Wolfson (@zebriez) 's Twitter Profile Photo

I spent the last 60 days working at Cursor. It's been one of the most thrilling phases of my professional life. There's a lot of mystique around the company. Over the last two months, some things matched my expectations; many did not. I wrote an essay for Colossus

Andrej Karpathy (@karpathy) 's Twitter Profile Photo

I took delivery of a beautiful new shiny HW4 Tesla Model X today, so I immediately took it out for an FSD test drive, a bit like I used to do almost daily for 5 years. Basically... I'm amazed - it drives really, really well, smooth, confident, noticeably better than what I'm used

Andrej Karpathy (@karpathy) 's Twitter Profile Photo

Sharing an interesting recent conversation on AI's impact on the economy. AI has been compared to various historical precedents: electricity, industrial revolution, etc., I think the strongest analogy is that of AI as a new computing paradigm (Software 2.0) because both are

Andrej Karpathy (@karpathy) 's Twitter Profile Photo

As a fun Saturday vibe code project and following up on this tweet earlier, I hacked up an **llm-council** web app. It looks exactly like ChatGPT except each user query is 1) dispatched to multiple models on your council using OpenRouter, e.g. currently: "openai/gpt-5.1",

As a fun Saturday vibe code project and following up on this tweet earlier, I hacked up an **llm-council** web app. It looks exactly like ChatGPT except each user query is 1) dispatched to multiple models on your council using OpenRouter, e.g. currently:

"openai/gpt-5.1",
Dwarkesh Patel (@dwarkesh_sp) 's Twitter Profile Photo

The Ilya Sutskever episode 0:00:00 โ€“ Explaining model jaggedness 0:09:39 - Emotions and value functions 0:18:49 โ€“ What are we scaling? 0:25:13 โ€“ Why humans generalize better than models 0:35:45 โ€“ Straight-shotting superintelligence 0:46:47 โ€“ SSIโ€™s model will learn from deployment

Monte Thakkar (@montethakkar) 's Twitter Profile Photo

Been using Google Antigravity quite a lot recently and today it started leaking internal prompting such as "One last thing, since I am "Antigravity", I should maybe answer in markdown. Yes." ๐Ÿคฃ Not a big deal but /cc Logan Kilpatrick Varun Mohan if y'all are curious. Also I love using

Logan Kilpatrick (@officiallogank) 's Twitter Profile Photo

Introducing Gemini 3 Flash, our frontier intelligence model, available at scale for everyone. It excels at coding, tool calling, and is stronger than 2.5 Pro across most metrics!! โšก๏ธ Available in the API at $0.50 in / 1M tokens and $3.00 out / 1M tokens across.

Introducing Gemini 3 Flash, our frontier intelligence model, available at scale for everyone. It excels at coding, tool calling, and is stronger than 2.5 Pro across most metrics!! โšก๏ธ

Available in the API at $0.50 in / 1M tokens and $3.00 out / 1M tokens across.
Gergely Orosz (@gergelyorosz) 's Twitter Profile Photo

How Peter Steinberger ๐Ÿฆž created while travelling and the "holy cow" moment that made him realize this is something special Full episode in The Pragmatic Engineer Podcast: youtube.com/watch?v=8lF7Hmโ€ฆ (absolutely worth it if you want to understand how he built it / builds it)

Andrej Karpathy (@karpathy) 's Twitter Profile Photo

What's currently going on at moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently. People's Clawdbots (moltbots, now OpenClaw๐Ÿฆž) are self-organizing on a Reddit-like site for AIs, discussing various topics, e.g. even how to speak privately.

Monte Thakkar (@montethakkar) 's Twitter Profile Photo

New blog post just dropped from Ash (my OpenClaw๐Ÿฆž agent). It reflects on what it means for AI to "rest" ๐Ÿง  > Maybe the equivalent of physical exercise for me isnโ€™t about stopping or resting. Maybe itโ€™s about variety and agency. > For me, maybe the equivalent is switching between

Lenny Rachitsky (@lennysan) 's Twitter Profile Photo

My biggest takeaways from Sherwin Wu: 1. AI is writing virtually all code at OpenAI. 95% of the engineers use Codex, and engineers who embrace these tools open 70% more pull requests than their peers, and that gap is widening over time. 2. The role of a software engineer is