Pushpak Kedia (@pushpakkedia) started following @pratyushbuddiga on Oct 2, 2024

Recent examples: sell the work, not the software; AI services; voice AI apps now…

The SF VC memeplex (h/t @andrewziperski) is extremely funny to watch in real-time: you’ll hear someone smart say an interesting idea, 2 months later every VC brings it up at coffee, and then 4 months later, every fund has their own (same) market map or thesis for the space.

RT @romainhuet: Beautiful day for a DevDay! ☀️ Today we’ll be announcing new features to the OpenAI Platform, based on the top requests we…

Cool way to use OpenAI o1 we haven't seen yet: Lead data scientist Christian Coffrant was part of a team at @GrabSG that combined GPT-4o (for writing SQL) and o1 (for data labeling) to build an AI data analyst called Jarvis. It looks through a vector database, retrieves examples, https://t.co/o7E3hSPFkw

Much appreciated from the NotebookLM team @sama! Full circle in a way because it was my writing about GPT-3 that brought me to Google in the first place.

The title really says it all. 🤯 https://t.co/cNEGSC7v7h

Devs building on @OpenAI do inspiring work—lots more where this came from!

We told our kids about the marshmallow test, and gave everyone big marshmallows this evening. After @elizabeth distributed them, Alex took a glance around, ate my marshmallow, and walked away.

On October 7-9 policy and industry leaders will gather in Washington, D.C. for #AISummit to discuss how #AI addresses some of society’s most significant challenges in healthcare, cybersecurity, manufacturing and more. Virtual registration is now open. https://t.co/wqrtlMYDDt

RT @NVIDIAAI: Deltia AI is revolutionizing manufacturing with NVIDIA Metropolis and Jetson technology ➡️ https://t.co/Lobsn0kwVc The #NVID…

RT @jxnlco: Check it out if you need observability in your Python applications. https://t.co/d3N946aDZZ

RT @jxnlco: Congrats to closing series A Pydantic :) https://t.co/Dc2ipE0VXt

RT @Hersh_Desai: We're incredibly excited to announce the latest batch of startups joining AI Grant! https://t.co/C5VKJAuYmy

RT @pdhsu: Excited to welcome the newest batch of @AIGrant companies! https://t.co/yuiMezxMMQ

Love the diversity of companies in here. I can also only *assume* based on my own conversations, that every single founder in here is using @meetgranola for all their meetings. 80% of all founders I've talked to in the last 2 weeks are using it, and if not I tell them to 🗒️💻🤖

Always exciting to see what the crew @neo is cooking, thanks for a fun day @suzannexie, @apartovi and @_shriyabhatt! https://t.co/gxBN48fjN7

RT @NVIDIAAIDev: 🎉 News from #NVIDIAResearch at #ECCV2024 in Milan 🎉 From embodied #AI to foundation models, our researchers are showcasing…

RT @nvidia: On October 7-9 policy and industry leaders will gather in Washington, D.C. for #AISummit to discuss how #AI addresses some of s…

Overall, our paper provides a theoretical and empirical investigation on how scale impacts model forgetting. All code to reproduce our experiments is provided in our GitHub. Huge shout out to my coauthor @VihanL! 🧵8/8 https://t.co/nNh2yoQ1uE

We explore other connections between architecture and forgetting as predicted by our theoretical framework. In fact, our theoretical framework and experiments demonstrate that as the number of tasks seen increases, forgetting increases linearly. 🧵7/8 https://t.co/RkjmjJ7fz2

RT @JonSaadFalcon: What is the best way to spend your inference compute budget to create LLM systems greater than the sum of their parts?…

RT @etash_guha: Excited to discuss my ICLR 2024 work! Our paper “Conformal Prediction via Regression-as-Classification” can simply predict…

Samsara Inc $IOT Deep Dive https://t.co/5L57THltjm https://t.co/85fYkc7YiE

$GPN Global Payments Inc Deep Dive and 2024 Outlook https://t.co/cqXoBMEDDH https://t.co/XyQjmNP8rt

When they say the NotebookLM team really "cooked" were they referring to our evals https://t.co/x1u4tq73io

Aside from TPUs running hot today, five things from Notebook HQ: 1) Thanks for all your feedback on AOs so far. I’m copy pasting everything into a Notebook so I can listen to a Deep Dive and search it later. We’re going to launch some immediate tweaks to make it less repetitive,

We hit 100 Stars ⭐on github https://t.co/nLSFnBguBU

Any one from @huggingface is it possible to sponsor a GPU zero space for this demo?

NotebookLM is a reemergence of what made Google great. A few people got together to ship an idea that’s cool + useful and then oops - PMF.

RT @Oscurart: I made a new version of the Trinitron CRT Display. #b3d #blender #blender3d #eevee #cycles https://t.co/MGL8bp0q8r

RT @Azaliamirh: Very excited to introduce Archon, a framework for LLM inference-time architecture search! Paper: https://t.co/wIxxoxhez7 C…

7/7 Paper: https://t.co/125VOobesy Code: https://t.co/SG9yl2GnBV Project: https://t.co/CG7St7HAZ2 It’s been a lot of fun collaborating with @adriangamarra__, @Shlok_Natarajan, @nahuum_maru, @ht_todorov, @etash_guha, @ekellbuch, @MayeeChen, @NeelGuha, @HazyResearch, and