Caithrin Rintoul

@willccbb @snowclipsed @samsja19 Read him the deep lore

@snowclipsed @samsja19 not a typo

anthoripic a prime intellect typo, close enough, welcome back @samsja19

Who is the Carl Sagan of AI? caithrin.substack.com/p/who-will-exp…

Facts

TBPN acquired by OpenAI 80,000 Hours Podcast acquired by Anthropic Nick Fuentes acquired by xAI

A ChinaTalk deep dive into Unitree's IPO chinatalk.media/p/unitrees-ipo

tbh I only feel more accelerationist as the capabilities ramp … the scaredest I was was in early 2023

ccunpacked.dev Learning from this, v cool

One of my favorite roundups of open models we've done, now on it's 20th month. There are a ton of diverse and interesting models across niches, rather than only more big MoE models for general use. Check it out & let us know what we missed!

remember when "12 days of Shipmas" was like, a really impressive amount of releases?

@bcherny @Ra1kshit @claudeai Why can't you just see how many tokens I burn on average per hour and give me a sense of how many hours of Claude usage I have? And then show me, if I used off-hours, how many more hours of work I could do. Current UX feels like punishment.

Working as hard as we can to make this better. It’s not easy growing at this rate, and it has been straining our services. Thanks for bearing with us. A number of significant improvements landed in the last few Claude Code releases and a few more on the way. Make sure you’re on the latest version.

Claude code is essentially unusable for me i hit 36% session limits in 15 mins. Just by using a single agent on my codebase. @claudeai you need to fix it. I'm on the 100$ plan btw

@jbhuang0604 Great work.

New video! How do LLMs grow outrageously large yet blazingly fast? The secret: Mixture of Experts (MoE) In this video, we cover the role of FFNs, how to scale them up without slowing down, and how to maintain load balance and stability. Full video here: youtu.be/0QQlYR1r6pQ

@soldni @allen_ai The finest of cogs! The kindest of cogs! The most generous of cogs!

@caithrin @allen_ai it was a team effort, i’m just one of the cogs 🫡

After 4yrs, today is my last day at @allen_ai It was an honor to work on Olmo, Dolma, olmOCR, Tulu, Molmo & other fully-open artifacts 🫡 Reception has been amazing & their adoption makes me SO PROUD 🥹 Team is super committed to open recipes; can't wait to see what's next!!!!

drake: i don't do K2, Kimi, deepseek that's more for your kind/my gyal more like demis, theme park, fat ting london deepmind meek mill: i claude coded percs outta 18th and berks / i got 500 agent lawyers tryna free lil durk drake singing: you got me loco tryna be ur sholtoOoO

@llm_wizard Cast this in bronze and put it in front of the building

And they call me…. Joe Nemotron.

@cjpedregal Yesss nice work!

There are some tweets out there saying that Granola is trying to lock down access to your data. Tldr; we are actually trying to become more open, not closed. We’re launching a public API next week to complement our MCP. Read on for context. A couple months ago, we noticed that some folks had reversed engineered our local cache so they could access their meeting data. Our cache was not built for this (it can change at any point), so we launched our MCP to serve this need. The MCP gives full access to your notes and transcripts (all time for paid users, time restricted for free users). MCP usage has exploded since launch, so we felt good about it. A week ago, we updated how we store data in our cache and broke the workarounds. This is on us. Stupidly, we thought we had solved these use cases well enough with our MCP. We’ve now learned that while MCPs are great for connecting to tools like Claude or chatGPT, they don’t meet your needs for agents running locally or for data export / pipeline work. So we’re going to fix this for you ASAP. First, we’ll launch a public API next week to make it easier for you to pull your data. Second, we’ll figure out how to make Granola work better for agents running locally. Whether that’s expanding our MCP, launching a CLI, a local API, etc. The industry is moving quickly here, so we’d appreciate your suggestions. We want Granola data to be accessible and useful wherever you need it. Stay tuned.

@JordanNanos congrats!

let’s go

@samt_granola @appenz @meetgranola Ahhh ok Drop me a DM I have some feedback on this!

@caithrin @appenz @meetgranola Claude pls 🥲 Mostly yes, we do some offline caching to make the app snappy but these are regularly pruned (also to keep the app snappy!)

Sorry to see Granola @meetgranola going closed. They encrypted their local db, no local and no cloud API. In a world where notes are managed by agents, the app now has zero value. Any recommendations for good alternatives? What are you switching to?

@samt_granola @appenz @meetgranola Claude tells me differently! Are they now stored server-side only?

@caithrin @appenz @meetgranola Transcripts are available in the MCP on Granola Business/Enterprise!