Maruti Agarwal

1K posts

Maruti Agarwal

Maruti Agarwal

@marutiagarwal

AI guy

Raleigh, NC Katılım Ağustos 2011
853 Takip Edilen712 Takipçiler
Base44
Base44@Base44·
Introducing Base44 Superagents. AI agents built with managed infrastructure, secured by default, one-click integrations, and 24/7 execution from the start. Everything is taken care of so you can focus on what your agent does, not how to get it running. That means no API keys to juggle, no config files, no security setup, and no maintenance. We handle all of it. Your Superagent connects to all the tools you already use in one click, runs on schedules and triggers, remembers context across sessions, acts proactively on your behalf, and keeps working around the clock. All from wherever you already are, WhatsApp, Telegram, Slack, or your browser. The AI agent everyone's been waiting for, with everything you need already built in. We're excited to get this into your hands, so we're giving free credits to everyone who comments and reposts in the next 24 hours.
English
1.1K
858
2.5K
1.6M
Maruti Agarwal retweetledi
Ilya Sukhar
Ilya Sukhar@ilyasu·
Asking a frontier closed model to prompt engineer a local small model is ... not working well.
English
3
1
6
1.4K
debayan
debayan@debayan·
In hindsight, that would be a good reason to come back to academia.
English
1
0
0
40
debayan
debayan@debayan·
Quitting academia because I am rich.
debayan tweet media
English
1
0
2
96
debayan
debayan@debayan·
Successfully defended my PhD thesis today.
debayan tweet media
English
153
135
5.1K
164.2K
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@debayan Lol… did I hear langchain :P… also way too many libraries to choose from and proportionally abandoned projects
English
0
0
0
78
debayan
debayan@debayan·
One problem with the rapid pace of development of AI is that the library APIs are changing at a faster pace than the documentation. I am often reading function descriptions which are now obsolete.
English
1
0
1
179
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@NirantK It’s strange to associate wealth with any of the other stuff you mentioned.
English
0
0
0
123
Nirant
Nirant@NirantK·
Because the last news you want to hear, is that this person who is wealthier than you, is also smarter, happier, and not a bad person morally!
English
3
2
29
3.6K
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@debayan I have seen cases where bunch of smarter combining strategies had a hard time beating majority voting :)
English
0
0
0
38
debayan
debayan@debayan·
I have been looking for papers which show tf-idf + linear SVM based classifiers perform worse than BERT based classifiers, but surprisingly, the ones I did find, present a mixed result. BERT is not clearly the winner.
English
1
0
1
217
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@agishaun I disagree. RAG not jsut about dumping data but also focus. Too much context thrown at LLM directly may also make response quality much worse. Would be harder to get to follow more than a few instructions. It might be useful for summarisation, sure. But replacing RAG, I doubt!
English
0
0
1
73
Shaun.AGI
Shaun.AGI@agishaun·
10M context window Gemini 1.5 is here. New tool in town! For long context window I usually have two questions, 1. How much does it charge? 2. How well does it perform compared to small context window? If 10M token window really works as expected, then no RAG would exist. You can dump pretty much 90% of your dataset.
Jeff Dean@JeffDean

Gemini 1.5 Pro - A highly capable multimodal model with a 10M token context length Today we are releasing the first demonstrations of the capabilities of the Gemini 1.5 series, with the Gemini 1.5 Pro model. One of the key differentiators of this model is its incredibly long context capabilities, supporting millions of tokens of multimodal input. The multimodal capabilities of the model means you can interact in sophisticated ways with entire books, very long document collections, codebases of hundreds of thousands of lines across hundreds of files, full movies, entire podcast series, and more. Gemini 1.5 was built by an amazing team of people from @GoogleDeepMind, @GoogleResearch, and elsewhere at @Google. @OriolVinyals (my co-technical lead for the project) and I are incredibly proud of the whole team, and we’re so excited to be sharing this work and what long context and in-context learning can mean for you today! There’s lots of material about this, some of which are linked to below. Main blog post: blog.google/technology/ai/… Technical report: “Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context” goo.gle/GeminiV1-5 Videos of interactions with the model that highlight its long context abilities: Understanding the three.js codebase: youtube.com/watch?v=SSnsmq… Analyzing a 45 minute Buster Keaton movie: youtube.com/watch?v=wa0MT8… Apollo 11 transcript interaction: youtube.com/watch?v=LHKL_2… Starting today, we’re offering a limited preview of 1.5 Pro to developers and enterprise customers via AI Studio and Vertex AI. Read more about this on these blogs: Google for Developers blog: developers.googleblog.com/2024/02/gemini… Google Cloud blog: cloud.google.com/blog/products/… We’ll also introduce 1.5 Pro with a standard 128,000 token context window when the model is ready for a wider release. Coming soon, we plan to introduce pricing tiers that start at the standard 128,000 context window and scale up to 1 million tokens, as we improve the model. Early testers can try the 1 million token context window at no cost during the testing period. We’re excited to see what developer’s creativity unlocks with a very long context window. Let me walk you through the capabilities of the model and what I’m excited about!

English
3
1
1
1.9K
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@jsappington I can make an introduction. DM me to whom and where you need the intro.
English
0
0
0
40
jacob sappington
jacob sappington@jsappington·
Who do I know that is a full on deliverability expert?
English
17
0
8
4.5K
Maruti Agarwal retweetledi
ShopAgain
ShopAgain@shopagainio·
As Valentine's Day approaches, the opportunity to connect with your customers and boost your sales is knocking at your door. Here are 11 Cupid-Approved Marketing Tips for Valentine’s Day Success in 2024 - buff.ly/4btt0ZW
ShopAgain tweet media
English
0
2
1
158
Nick Huber
Nick Huber@sweatystartup·
Walk into every single interaction with the goal of adding as much value as humanly possible. Don't think about charging money, withholding info, etc. Prove you know your stuff. Help in a massive way. Gain trust. Profit later.
English
4
21
405
32K
Nick Huber
Nick Huber@sweatystartup·
A mindset shift that really changed my life: A thread:
English
22
57
663
354.8K
debayan
debayan@debayan·
I love @MixtralAI ! On a sentence classification task, the 8x7b model's zero-shot accuracy matches that of GPT-4!
English
3
0
4
228
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@nachovanzini @obviceo Hmm… I have hired developers from LATAM in past… will check my connections if I can find something there
English
1
0
0
37
Nacho Vanzini | DTC Growth
Nacho Vanzini | DTC Growth@nachovanzini·
@obviceo There’s a ton of solid talent in Uruguay. Extremely high English proficiency and only 1-2 hrs ahead of EST
English
1
0
2
654
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@debayan The board might be too cluttered to play and it will never unclutter :P Why did she want more pieces?
English
0
0
0
28
debayan
debayan@debayan·
My daughter suggested two new rules to chess today. 1. Three rows of pieces instead of 2. 2. Pieces do not eat each other. They coexist on the same square.
debayan tweet mediadebayan tweet media
English
2
0
7
583
Maruti Agarwal
Maruti Agarwal@marutiagarwal·
@sweatystartup The way I see it is that building startup is also not a healthy for most people ;) but those who find it enjoyable, do it anyways. That’s the case with other extreme sports too!
English
0
0
0
29
Nick Huber
Nick Huber@sweatystartup·
I’m not sure I’ll let my kids run cross country. In my uneducated opinion: Competitive long distance running is not good for a human. Cycling is the same. These folks look 10-20 yr older than they are. Many have eating disorders and it is terribly sad - especially prevalent in the young ladies. Skin and hair look unhealthy on the folks who run really long distances competitively. Worse than football in my humble opinion… Am I wrong? What are your thoughts?
English
632
23
708
711.3K

Keşfet

@Base44 @debayan @NirantK @jsappington @agishaun @elonmusk @BarackObama @taylorswift13