Google, every mention
194 scenes (2024), the whole family · ← back to Google
tap a year for its mentions
every year 2024 anyone Shawn Wang 216Alessio Fanelli 62Dylan Patel 56Yi Tay 42Will Bryk 24Bret Taylor 19Jeremy Howard 18Omar Sanseviero 16Anjney Midha 16Varun Mohan 15
Verbatim, from the transcripts: the passages where Google comes up
2024 in Post-Transformer Architectures: State Space Models, RWKV [Latent Space LIVE! @ NeurIPS 2024]
- ▶ 0:59 Dan Fu Over the last five to six years, we've seen models really scale up in parameter size, and that's brought with it a bunch of new capabilities, like the ability to talk to you and tell you, um, sometimes how to use your Colab and your AWS…
- ▶ 24:19 Eugene Cheah Okay, because we are all GPU poor, um, and to be clear, like, most of this research is done, like, only on a handful H 100, which I had one Google researcher told me that was like his experiment budget for a single researcher. 2 times in the scene
- ▶ 31:55 Dan Fu What if we could remove that and get the same quality, but a lot faster generation time, um, or some of the demos that we saw from Page earlier today, you know, if, if I have a super long, uh, conversation with, with my Gemini bot, um,…
- ▶ 37:50 Eugene Cheah So, I'm going to burn a bit, uh, how many of you, uh, remember the news of Google Gemini supporting three million contacts, right? 2 times in the scene
Best of 2024: Open Models [LS LIVE! at NeurIPS 2024]
- ▶ 28:08 unnamed speaker So you can use our model on Google Cloud, AWS, Azure, uh, Snowflake, IBM.
The State of AI Startups in 2024 [LS Live @ NeurIPS]
- ▶ 3:38 Pranav Reddy Google had just announced Gemini. 2 times in the scene
- ▶ 3:40 Pranav Reddy I very genuinely forgot about the existence of Bard before making these slides.
Windsurf: The Enterprise AI IDE
- ▶ 4:42 Varun Mohan Like, when you use Google, do you say, like, at New York Times post
- ▶ 57:37 Varun Mohan It was like, Google's a great company, makes a lot of money. 2 times in the scene
- ▶ 57:39 Varun Mohan What happens if they actually made the search index of the product something that someone else bought for them? 2 times in the scene
0 to over $8M ARR in 2 months as a Claude Wrapper (Bolt.new, Qodo)
- ▶ 20:08 Eric Simons You go Google, like, uh, you know, turn Docker container into Wasm. 3 times in the scene
- ▶ 22:35 Eric Simons Everyone at the time, like we talked to people that like worked on, you know, uh, Chrome and VA and they're like,
- ▶ 34:54 Itamar Friedman The other could be, like, if it's air-gapped completely, or you want VPC, and then you have Azure, GCP, and AWS, which is different.
- ▶ 40:07 Itamar Friedman We see cases where you're running on GCP and they want open AI, like this cross, like a case. 2 times in the scene
- ▶ 1:03:56 Eric Simons So think like, kind of like Google CoLab sort of thing, or like Hugging Face has their kind of version of this.
[Paper Club] Embeddings in 2024: OpenAI, Nomic Embed, Jina Embed, cde-small-v1 - with swyx
[Paper Club] DocETL: Agentic Query Rewriting + Eval for Complex Document Processing w Shreya Shankar
- ▶ 44:50 Eugene Yan If you search it up on Google, I'm sure you will leave it to the archive.
The new Claude 3.5 Sonnet, Computer Use, and Building SOTA Agents — with Erik Schluntz, Anthropic
- ▶ 51:47 Erik Schluntz I mean, after the first, just like, you know, go to Google, I think I tried to have it play Minecraft or something 2 times in the scene
- ▶ 1:08:05 Erik Schluntz Uh, there's, um, Eric Jang was at, uh, he did AI at Google.
[Paper Club] BERT: Bidirectional Encoder Representations from Transformers
- ▶ 2:52 unnamed speaker Just for a little bit more context, so after, so this was a, a Google paper, and after they trained and released BERT, they started using it in, um, Google search, and so it provided context for search results so that 4 times in the scene
- ▶ 2:52 unnamed speaker Just for a little bit more context, so after, so this was a, a Google paper, and after they trained and released BERT, they started using it in, um, Google search, and so it provided context for search results so that
- ▶ 20:28 unnamed speaker Like Google doesn't need to spend millions of dollars to predict which sentence comes before or after another sentence, but it does help a small model better better learn embedding per se and abstract that to these tasks.
- ▶ 32:39 unnamed speaker So it was in the realm of like at the time, not as effective to get, you know, like you don't have Google compute to retrain Bert from scratch, but now there's stuff of like 24 hours and a couple hundred dollars and retrain your own better…
- ▶ 42:12 unnamed speaker Oh, it was well more than a 10 K project that Google, uh, the, the BERT large itself was already a 50 K train run plus 10 to 15 for BERT space plus just experimentation. 2 times in the scene
Agents @ Work: Lindy.ai (with live demo!)
- ▶ 6:05 unnamed speaker Uh, you have Slack, you have Google Calendar, you have Gmail.
- ▶ 6:47 Florent Crivello But for example, for Google, obviously there's hundreds of scopes that you could require for Google. 2 times in the scene
- ▶ 7:43 Florent Crivello And so in particular, Google Workspace accounts. 2 times in the scene
- ▶ 21:53 Florent Crivello But I think that Google, right? 4 times in the scene
- ▶ 22:18 Florent Crivello Like Google is a big company.
Agents @ Work: Dust.tt — with Stanislas Polu
- ▶ 2:09 Stanislas Polu At the time, I think a few years after it was the first project at Google, uh, but you know, that, that cat face or the human face, uh, trained from many images.
- ▶ 8:21 Shawn Wang There's a bit of work with, like, Lean, and then with, uh, you know, more recently with, uh, with DeepMind doing, like, scoring, like, Silver on the IMO. 2 times in the scene
- ▶ 13:54 Stanislas Polu What has been useful to the team and, and with the DeepMind results shows that it was the right approach.
In the Arena: How LMSys changed LLM Benchmarking Forever
- ▶ 23:12 Alessio Fanelli We also had Nicholas Carlini from the mind on the podcast.
[Paper Club] Upcycling Large Language Models into Mixture of Experts
How NotebookLM Was Made
- ▶ 1:33 Shawn Wang You know, what, what is your path into the sort of Google AI org, or maybe I actually don't even know what org you guys are in. 6 times in the scene
- ▶ 1:42 Raiza Martin I lead the Notebook LM team inside of Google Labs. 4 times in the scene
- ▶ 1:54 Raiza Martin We work super closely with DeepMind.
- ▶ 4:00 Shawn Wang May, 23, during Google I.O., you announced Project Tailwind. 5 times in the scene
- ▶ 4:36 Raiza Martin And that was the first project that I was working on at Google.
- ▶ 7:39 Raiza Martin So I think from the initial demo, I think there was so much excitement of like, wow, what is this thing that Google is launching? 2 times in the scene
- ▶ 8:07 Raiza Martin But then we basically took the feedback from IO.
- ▶ 9:33 Shawn Wang It's not like, it's not labs.
- ▶ 14:28 Usama Shafqat Like Ryza said earlier, like we work with the DeepMind audio folks pretty closely.
- ▶ 18:34 Raiza Martin And then he joined Google
- ▶ 22:18 Shawn Wang And he has notes, Google drive stuff, PDFs, MP three, whatever.
- ▶ 28:55 Raiza Martin I, um, I hear from Googlers all the time, especially because we launched it internally first. 3 times in the scene
- ▶ 42:36 Usama Shafqat I mean, I was literally chatting with someone at Google earlier today about how some people think that like you need a linguistics person in the room for like making a good chatbot, but that's not actually true because like this person…
- ▶ 56:26 Usama Shafqat The speech team supports like a bunch of other efforts across Google, like Gemini live.
- ▶ 56:30 Usama Shafqat For example, is also the models built by the same like sort of deep mind speech team. 2 times in the scene
- ▶ 1:02:38 Alessio Fanelli Any thoughts on some of the alternative interact with data and documents things, like cloud artifacts, like a JGBD canvas, you know, kind of how do you see maybe where Notebook LM starts, where like Gemini starts, like you have so many…
- ▶ 1:09:44 Raiza Martin Occasionally I'll, I'll try to build something on like GCP over the weekend.
- ▶ 1:10:52 Alessio Fanelli It's funny because we had a Nicola Scarlini from DeepMind on the pod 2 times in the scene
- ▶ 1:11:15 Raiza Martin Like I remember the first time I used Google search, I was like, what do we type?
- ▶ 1:12:14 Shawn Wang I think that notebook.lm was kind of like the ChatGPT moment for Google.
Singapore: the AI Engineer Nation — with Minister Josephine Teo
- ▶ 39:43 Shawn Wang Let's go and get Anthropik, Google DeepMind, all these, all these guys, the AI creators to move to Singapore.
Building the Silicon Brain - Drew Houston of Dropbox
- ▶ 8:07 Drew Houston And so, and then you think of other level one, level two type things like Google Maps probably did more for self-driving than than literal self-driving.
- ▶ 15:24 Alessio Fanelli I built a kind of like email triage assistant and like, I would say, 80% of the code is like Google or auth and like pulling the emails and then the actual AI part is.
- ▶ 21:32 Drew Houston Then on the other side of your screen, you have, like, Chrome or a browser that has so many tabs open, you can no longer see text.
- ▶ 23:20 Alessio Fanelli I think it's counterintuitive that in Dash you can connect Google Drive, right?
- ▶ 33:20 Alessio Fanelli I think today it's almost like, hey, if I can use Dash to like access my Google Drive file, why would I pay Google for like their AI feature?
- ▶ 33:20 Alessio Fanelli I think today it's almost like, hey, if I can use Dash to like access my Google Drive file, why would I pay Google for like their AI feature?
- ▶ 33:43 Drew Houston Well, I think you have to get the value exchange right, and I think you have to be, like, a trustworthy partner, or, like, no one's gonna partner with you if they think you're gonna eat their lunch, right, or they'll, if you're gonna… 2 times in the scene
- ▶ 34:52 Drew Houston So we're not really building like a copilot or Gemini competitor.
- ▶ 40:52 Drew Houston In IT, whose jobs it is to, like, manually go through, log into each, like, log into Office, log into Workspace, log into each tool, and, like, go comb through, one by one, the links that people have shared, and, like, unshare.
- ▶ 1:05:17 Drew Houston Then two years, it'll be like, well, we're gonna have this, like, product, but then we're gonna need users, we're gonna need money, revenue, and then in five years, it'll be like, yeah, we're gonna be, like, tangling with, like, Microsoft,…
[Paper Club] Molmo + Pixmo + Whisper 3 Turbo - with Vibhu Sapra, Nathan Lambert, Amgadoz
- ▶ 7:06 Vibhu Sapra It's a very Apple intelligence or Google lens type example, where I think the team put it on the Apple vision pro
- ▶ 15:46 Nathan Lambert I think there's, like, hundreds of people that filled out the silly Google form asking for interest, which I'm not surprised.
Production AI Engineering starts with Evals
- ▶ 53:08 unnamed speaker Uh, Google Sheets to custom tool.
- ▶ 1:05:24 unnamed speaker Yeah, and for those who don't know, Malta used to work on Google Search.
- ▶ 1:28:28 unnamed speaker They all use Azure for OpenAI because they have that special relationship, and then Google has Google. 2 times in the scene
Building AGI in Real Time (OpenAI Dev Day 2024)
- ▶ 1:03:15 Simon Willison Like Google Gemini, you can do an hour's worth of video in their context window, and just by slicing it up into one frame at 10 frames a second.
- ▶ 1:49:39 Sam Altman Um, I think Google's notebook thing is super cool.
Language Agents: From Reasoning to Acting — with Shunyu Yao of OpenAI, Harrison Chase of LangGraph
- ▶ 8:34 Shunyu Yao So the change came when I started the internship in Google and apparently Google care less about text game. 2 times in the scene
- ▶ 35:23 Shawn Wang Do you have thoughts on, uh, alpha proof that the new DeepMind paper?
- ▶ 49:26 Shawn Wang Like, can we just use Bing and Google for LLM search? 2 times in the scene
- ▶ 59:56 Harrison Chase It's on this Google Scholar.
- ▶ 1:01:22 Shunyu Yao So when I wrote down the React thing, I would tell it to my Google manager, right? 2 times in the scene
- ▶ 1:16:32 Shunyu Yao Does it show on Google?
llm.c's Origin and the Future of LLM Compilers - Andrej Karpathy at CUDA MODE
The Ultimate Guide to Prompting - with Sander Schulhoff from LearnPrompting.org
- ▶ 5:59 Sander Schulhoff So over the course of about nine months, I led a 30 person research team with people from OpenAI, Google, Microsoft, Princeton, Stanford, Maryland, a number of other universities and companies.
- ▶ 28:53 Shawn Wang That's a deep mind one.
- ▶ 28:54 Shawn Wang That's from Google.
- ▶ 59:52 Shawn Wang Google researchers about this, because they're working on video models as well, and we'll see what, we'll see what happens.
Building AGI with OpenAI's Structured Outputs API
- ▶ 0:42 Shawn Wang So you've interned and always worked at Google, Stripe, Coinbase, Clubhouse, and obviously OpenAI. 2 times in the scene
- ▶ 42:58 Alessio Fanelli We had one episode with, um, Nicola Scarlini from DeepMind, and we actually talked about how some people don't actually get to the
- ▶ 1:07:20 Shawn Wang The, the thing that we just talked about, MD-F crypt, Ruby on Rails, Google Maps, and V-Eight for, for Chrome.
- ▶ 1:07:20 Shawn Wang The, the thing that we just talked about, MD-F crypt, Ruby on Rails, Google Maps, and V-Eight for, for Chrome.
Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
- ▶ 25:48 unnamed speaker Specifically because you brought up spreadsheets, I want to share like my personal experience because I think Google's done a really good job that people don't know about, which is 2 times in the scene
- ▶ 25:55 unnamed speaker If you use Google Sheets, it's, Gemini's integrated inside of Google Sheets, and it helps you write formulas. 3 times in the scene
- ▶ 26:04 unnamed speaker Like, if you, unless you watch Google I.O., there was no other opportunity to learn that Gemini is now in your Google Sheets.
- ▶ 57:32 Nicholas Carlini I'm at Google. 3 times in the scene
Is finetuning GPT4o worth it?
- ▶ 4:34 Alistair Pullen And I then went home and like Googled GPT-III
- ▶ 23:21 Alistair Pullen I think obviously the most important tools to these agents are like being able to retrieve code from a code base, being able to read Stack Overflow articles and what have you, and just be able to essentially be able to Google like we do is…
Answer.ai & AI Magic with Jeremy Howard
- ▶ 14:52 Jeremy Howard You know, multiple of Stanford, Meta, DeepMind, OpenAI, Berkeley, Oxford. 2 times in the scene
- ▶ 23:57 Jeremy Howard He ran browsing and retrieval for, for Google's DeepMind stuff, um, created Gemma.cpp, and he's been working on a new, uh, system to make it easier to do web GPU programming.
- ▶ 58:48 Jeremy Howard So it's also the way Bart and Gemini work, it's also the way the Claude web app works.
Segment Anything 2: Memory + Vision = Object Permanence — with Nikhila Ravi and Joseph Nelson
- ▶ 1:29 Nikhila Ravi And in my machine learning class in undergrad, I remember one day our professor came in and that was when Google acquired DeepMind.
- ▶ 1:29 Nikhila Ravi And in my machine learning class in undergrad, I remember one day our professor came in and that was when Google acquired DeepMind.
page 1 of 2 · 100 scenes per page · newest episode first next →