Steve Messina @steve_messina
Co-founder/CTO https://t.co/4tZyFxCQxu , @ai_azara, ex-IBM CTO, ex-AWS Principal Architect, Devops & Containers inventor, ex Banking Chief Architect. personal account. azara.ai Singapore Joined March 2011-
Tweets26K
-
Followers962
-
Following2K
-
Likes119K
Let's learn how to build a financial agent that not only answers pointed questions over documents, but can perform complex computations on top of the tabular data within these docs 🧮 This requires the following layers on top of RAG: - a memory layer to track state (extracted…
Let's learn how to build a financial agent that not only answers pointed questions over documents, but can perform complex computations on top of the tabular data within these docs 🧮 This requires the following layers on top of RAG: - a memory layer to track state (extracted… https://t.co/sw5zkNK1d4
#OnThisDay in 1993, CERN put the World Wide Web in the public domain, later made available with an open licence, allowing the web to flourish. Here we see Sir Tim Berners-Lee's proposal for the World Wide Web, revised back in May 1990. The World Wide Web will be one of the…
BREAKING: @Tesla has secured a new ~$650 million contract to build the largest battery energy storage system in Australia, over twice the size of the storage system pictured below. It will mean the Neoen Collie battery will be sized at a total 560 MW & 2,240 MWh. This $650M…
Llama 3 degrades more than Llama 2 when quantized. Probably because Llama 3, trained on a record 15T tokens, captures extremely nuanced data relationships, utilizing even the minutest decimals in BF16 precision fully. Making it more sensitive to quantization degradation.…
Nvidia CEO: people with really high expectations have very low resilience. Success in life requires a high degree of resilience, grit, hard work and optimism.
The most concerning part about this is that most Europeans don't realize how stagnant Europe has now become Europeans are literally blue-pilled and are mostly concerned with climate change, immigration and the Ukraine war Nobody in Europe is thinking why increasingly everything…
The most concerning part about this is that most Europeans don't realize how stagnant Europe has now become Europeans are literally blue-pilled and are mostly concerned with climate change, immigration and the Ukraine war Nobody in Europe is thinking why increasingly everything… https://t.co/Y68yzRfgKj
In 2024, we should see some incredible innovation around small language models. This new paper has nice idea in that direction. "Blending Is All You Need: Cheaper, Better Alternative to Trillion-Parameters LLM" 📌 "when specific smaller models are synergistically blended, they…
Untrue. This does touch on something related that actually happened, which people have apparently distorted and used to prop up the dumb conspiracy theory. I will explain… 1/N
Untrue. This does touch on something related that actually happened, which people have apparently distorted and used to prop up the dumb conspiracy theory. I will explain… 1/N
🧠 Run LLama 3 70B on a Single 4GB GPU - with airllm and layered inference 🔥 layer-wise inference is essentially the "divide and conquer" approach 📌 And this is without using quantization, distillation, pruning or other model compression techniques 📌 The reason large…
🤖AutoHyDE: Making HyDE Better for Advanced LLM RAG A fun article exploring new and creative RAG techniques Builds upon HyDE to add in keyword generation, clustering, etc Implements a LangChain Embedder which makes it easy to try out in your pipeline towardsdatascience.com/autohyde-makin…
“So I asked Ilya Sutskever, OpenAI’s chief scientist, for a reading list. He gave me a list of like 40 research papers and said, ‘If you really learn all of these, you’ll know 90% of what matters today.’ And I did. I plowed through all those things and it all started sorting out…
Based on the new changes to OpenAI Assistants, here's what Enterprises have been asking for: - Enterprise-grade security - Better administrative control - More "knowledge" (files) - Streaming support - Ability to choose specific tools - Support for fine-tuned 3.5 models - Lower…
Another great Langchain🦜 x CopilotKit🪁 tutorial Check it out to get a glimpse into the future of productivity applications
Another great Langchain🦜 x CopilotKit🪁 tutorial Check it out to get a glimpse into the future of productivity applications
llmlingua - This great lib from Microsoft can 𝗰𝗼𝗺𝗽𝗿𝗲𝘀𝘀 your prompt massively. 📌 Up to 20% of the prompt's original length (5x reduction), leading to massively reduced cost and latency. 🔥 speed up LLMs' inference and enhance LLM's perceive of key information, compress…
In the weightlessness on @Space_Station I placed a sphere of water with an air bubble on a speaker cone, then drove it with acoustical waves from the speaker. This created a series of standing nodes and anti-nodes that responded to the frequency and amplitude. A zero gravity…
Tesla Has 400 Exaflops of AI Training Compute dlvr.it/T5ly6R
John Arundel @bitfield
98K Followers 100K Following Go/Rust mentor and author, 'Explore Go: Cryptography' and other stories. Programming is fun, and you should have fun. 🦀Baslushi @baslushi75156
0 Followers 209 FollowingSpairs @Spairs1147
0 Followers 211 FollowingOlgaKipling @2Ew4w0rz0f5Hf
0 Followers 120 FollowingDartought @DartoughtDSgFt
0 Followers 90 FollowingCootosm @cootosm140
45 Followers 2K FollowingSangmin Bae @raymin0223
443 Followers 650 Following Ph.D. candidate at @kaist_ai; Efficient NLP, Self-Supervised Learning, Federated Learning, Generative AI, and Multimodal LearningSiyi Zhang @ameliadee20_
1K Followers 23 Following Always remember that in life. Whatever we do, we are never defeated unless we give up.USAchewingCode @chewingcode
1K Followers 395 Following software engineer; https://t.co/qWiFmjqMOj; front-end, half back-end, one third designer;Facteur @FacteurDev
278 Followers 2K Following An open-source alternative to @Resend. A @ValyentDev initative.simpletrading @simpletrad17722
425 Followers 7K FollowingGenevieveOwen @CG6zt9hP1vIN6kq
0 Followers 137 FollowingSewhayc @sewhayc30661
44 Followers 2K FollowingHenry Jin @jinnyboiii
1K Followers 4K Following Building @withLeoAI | AI Toolkit for Teachers https://t.co/4SSkxVltTn @Harvard and @UCBerkeley alumElon Musk @elonmusk_1909
142 Followers 4K FollowingMarcus Canhaço @CanhacoMarcus
2 Followers 38 FollowingChuong M. Huynh @RyanHuynh1108
282 Followers 475 Following Incoming Research Intern at @AmazonScience; Ex-Research Intern at @AdobeResearch; Formerly @VinAI_Research; Ph.D. student @UMDCSRæ Rosé @aLayLampada
30 Followers 600 FollowingHyeonbin Hwang @ronalhwang
148 Followers 206 Following M.S. Student @kaist_ai https://t.co/bQW6mlGzDNiuri matias @iurimatias
5K Followers 377 Following AI & Decentralization Enthusiast | Head of Desktop at @ethstatus formerly: Remix Core Dev @ethereum, Lead Dev @embarkprojecttoni breidinger @toni_1830
440 Followers 5K Following Massage therapist Level 3 FST -Practitional 8 Detoxification~Relaxation -Rejuvenation Specialized In All Kind Of Massage 5 Send a DM to Book. money matters💚James Murdoch 𝕏 @jamesmurdooch
1K Followers 2K Following TOP BOARD MEMBER OF TESLA INC ,SpaceX , quantum Ai (Mars Exploration)Lime Rick @LRCRB_LDL
824 Followers 4K Following Euclids*DNA\Claudius×Schweitzer/Runners°Resurrection/window0wonday/Whatapersonification&The/Latter/Day/Leo/\de\1984+10\AKA/the\Lime/Rick\On/A\Corona/Roll-bi*chearf0990 @P3rkins
79 Followers 2K FollowingChris Latimer @chrislatimer
1K Followers 490 Following Co-founder & CEO @Vectorizeio. ex-Google. Helping developers and companies accelerate gen AI application development.Kai-Fu Lee @kaiifulee
1K Followers 2K Following #AI Expert, CEO of @01ai_yi and Chairman of 创新工场 @sinovationvc , former President of Google China, Author of AI 2041 and NYT Bestseller AI Superpowerswintermew @wintermew121962
28 Followers 2K FollowingKimberly @yoshihi98515818
3 Followers 794 Following钟辉 @zhnghu008049148
8 Followers 342 FollowingSmestepesh @smestepesh84003
29 Followers 2K Following三上 春香 | JINGS.. @haru__mika
658 Followers 541 Following JINGS CEO | MACC所属 🪷企業マーケ領域のAIツールを作ってます!(開発受付中) エンジニア積極採用中Juan Pablo Mesa Lopez @juanpml_
79 Followers 423 Following AI Engineer Focused on LLMs, RAG, Multimodal embeddings, and building AI-powered software development. Sharing my real-world experiences and learnings in AI.follow @follow880761
362 Followers 5K FollowingConsiglieri @ConsiglieriVita
85 Followers 2K FollowingVijay Mandava @mandava_vijay
9 Followers 120 FollowingPaloma Jimenez @palomajimenez45
516 Followers 2K Following Manifesting Prosperity $ #I Spreading Wealth Affirmations & | Empower your journey to AbundanceMichael Liebmann @LiebmannMichael
49 Followers 148 Following Serial Founder | PhD in AI 🤖 | Building the tech framework for an innovative & sustainable species 🌍 | Believer in free data | Let's evolve together. 💡Bogdan Beldiman @bogdanonymous
867 Followers 4K Following Self-Educated Multidisciplinary Experimenter | AI EnthusiastSiva @sivaram2k10
627 Followers 4K Following #Solopreneur - #AI - #Image, #Text, #Data. #DataAnalytics, #Advisory, #Teaching, Product development. Open for AI ML product development / collaborationThealelos @thealelos73994
27 Followers 2K FollowingValts Ausmanis @ValtsAusmanis
250 Followers 321 Following AWS Cloud Architect ☁️ ‧ Software Developer ‧ Co-founder of @CloudvizIoJohn Arundel @bitfield
98K Followers 100K Following Go/Rust mentor and author, 'Explore Go: Cryptography' and other stories. Programming is fun, and you should have fun. 🦀GitHub @github
2.6M Followers 341 Following The AI-powered developer platform to build, scale, and deliver secure software.Phil (Newsletter link.. @BadAstronomer
592K Followers 480 Following SCIENCE! I love this stuff. Sign up for my Bad Astronomy Newsletter! Link is below. I use likes for bookmarks; no endorsement implied. He/him.Gergely Orosz @GergelyOrosz
249K Followers 2K Following Writing @Pragmatic_Eng, the #1 technology newsletter on Substack. Author of @EngGuidebook. Formerly Uber & Skype.Chris Hadfield @Cmdr_Hadfield
2.3M Followers 200 Following Astronaut, back on Earth after 3 spaceflights. For events and media, please write to [email protected] Order books at https://t.co/Y7p3sa0MX3@levelsio @levelsio
419K Followers 1K Following 🧢https://t.co/w98s8lFJiK $5K/m 💆https://t.co/AoNP9BW2Dp $3K/m ✨https://t.co/BmbkrX4Zyf $0K/m 📸https://t.co/lAyoqmSBRX $57K/m 🖼https://t.co/1oqUgfD6CZ $44K/m 🌍https://t.co/BjTozWAXwG $27K/m 🛰https://t.co/ZHSvI2wjyW $51K/mLutra AI @Lutra_AI
709 Followers 7 Following Automate your work with AI. Lutra is the future of automation. Create AI workflows just from English instructions without the need for coding or drag-and-dropFilip Pyrek 🍬 is b.. @FilipPyrek
2K Followers 990 Following AWS Serverless Hero 🚀 / Founder of @buttonizeio 🍬 / Serverless Meetup Co-Organizer 🇨🇿 / https://t.co/456jQah5peSangmin Bae @raymin0223
443 Followers 650 Following Ph.D. candidate at @kaist_ai; Efficient NLP, Self-Supervised Learning, Federated Learning, Generative AI, and Multimodal LearningNayan Saxena @SaxenaNayan
2K Followers 2K Following Brought artificial intelligence to @RBC, @Glowforge, @Wombo, @Bell & beyond.LLMOps Space @llmopsSpace
1K Followers 1K Following https://t.co/BsJFeDnBaD is a global community for LLM practitioners. 💡📚Facteur @FacteurDev
278 Followers 2K Following An open-source alternative to @Resend. A @ValyentDev initative.Aaron Gokaslan @SkyLi0n
3K Followers 345 Following Creator of the OpenWebText and OpenGPT2. @PyTorch Core Reviewer. PhD Student at @Cornell (interning at @MosaicML) Previously at @FacebookAI and @BrownUniversityTheSpaceEngineer @mcrs987
6K Followers 369 Following 3d artist, software engineer & self-proclaimed propulsion tech nerd @RGVaerialphotos & @LabPadre VFX team @IntegrationGame development personal account:@jnm1337chewingCode @chewingcode
1K Followers 395 Following software engineer; https://t.co/qWiFmjqMOj; front-end, half back-end, one third designer;Jeremiah Owyang @jowyang
251K Followers 984 Following Venture Capital Investor at Blitzscaling Ventures @BlitzVentures | Llama Lounge organizer & Conference Speaker | Works in an Airstream | CrossFitHyeonbin Hwang @ronalhwang
148 Followers 206 Following M.S. Student @kaist_ai https://t.co/bQW6mlGzDNChris Stanchak @chrisstanchak
5K Followers 634 Following Starter of Things | Cofounder/CEO Loveseat. Previous Founder/CEO TicketLeap. Advisor Venmo. Wharton 🎓Chris Latimer @chrislatimer
1K Followers 490 Following Co-founder & CEO @Vectorizeio. ex-Google. Helping developers and companies accelerate gen AI application development.Weaviate • vector d.. @weaviate_io
12K Followers 3K Following The easiest way to build and scale AI applications. 🐙 https://t.co/9ZP8iC4iFd 📰 https://t.co/XiFW3Ks5fKJonathan Tuite @weaviate_jon
25 Followers 2 FollowingJuan Pablo Mesa Lopez @juanpml_
79 Followers 423 Following AI Engineer Focused on LLMs, RAG, Multimodal embeddings, and building AI-powered software development. Sharing my real-world experiences and learnings in AI.Sam Whitmore @sjwhitmore
12K Followers 2K Following building @newcomputer. not a cat (or a man) in real life. I like to run a lot! @kensho @harvard @StuyNYHolgerJohnVogel @HolgerJohnVogel
109 Followers 275 Following Financial Digital Engineer on Structured Dematerialized Commodity - LBMA like 'Good Delivery' AI Code made within Specifications Terms and ConditionsWenhu Chen @WenhuChen
11K Followers 520 Following AI researcher @UWaterloo @GoogleAI @VectorInst. Interested in natural language processing, diffusion models. I direct TIGER-Lab at UWaterloo.falkordb @falkordb
127 Followers 44 Following The first queryable Property Graph database to use sparse matrices to represent the adjacency matrix in graphs and linear algebra to query the graph.Philipp Schmid @_philschmid
16K Followers 653 Following Tech Lead and LLMs at @huggingface 👨🏻💻 🤗 AWS ML Hero 🦸🏻 | Cloud & ML enthusiast | 📍Nuremberg | 🇩🇪 https://t.co/l1ppq3q3hkCasper Hansen @casper_hansen_
2K Followers 175 Following NLP Scientist | AutoAWQ Creator | Open-Source ContributorJosephine Chen @josephinekchen
3K Followers 329 Following Partner @Sequoia (seed/A/B). Chef @joinRamenMafiaDmitry Noranovich @javaeeeee1
1K Followers 3K Following Software developer with physics background, teacher, entrepreneurYoshiki Miura @miiura
4K Followers 819 Following AI Product Engineer and Founder of Morphic Morphic: A fully open-source AI-powered answer engine with a generative UI BabyAGI UI: An open-source UI for BabyAGIMichael Liebmann @LiebmannMichael
49 Followers 148 Following Serial Founder | PhD in AI 🤖 | Building the tech framework for an innovative & sustainable species 🌍 | Believer in free data | Let's evolve together. 💡Jordan Burgess @jordnb
5K Followers 2K Following cofounder and chief prompt officer @humanloop — manage and evaluate your LLM applications🔥 Matt Dancho (Bus.. @mdancho84
54K Followers 464 Following On a mission to grow your business data science skills and accelerate your career | Get my free 5-day business for data scientists course 👇Alexander Nevedovsky @ednevsky
500 Followers 2K Following AI entrepreneur. Ex WANNA (acquired by Farfetch) and Palta (300M users). Advising/investing in founders via @fund_hackers 🛠️ML Collective @ml_collective
13K Followers 41 Following We are an independent nonprofit organization that believes collaboration opportunities and research training should be openly accessible and free.Andrew Nguonly @andrewnguonly
291 Followers 634 Following swe @LangChainAI | writing @medium | prev @southpkcommons @netflixValts Ausmanis @ValtsAusmanis
250 Followers 321 Following AWS Cloud Architect ☁️ ‧ Software Developer ‧ Co-founder of @CloudvizIoCodeRabbit @coderabbitai
9K Followers 19 Following Supercharge your dev team with AI-driven contextual code reviews & smart chat https://t.co/HfgoeB5HPQAgentOps 🖇️ @AgentOpsAI
3K Followers 5 Following Agents suck. We're fixing that. (DMs open). https://t.co/GcfVtDtYlZVarun @varun_mathur
24K Followers 7K Following Make AI Free. Co-founder and CEO @HyperspaceAI. Run a node: https://t.co/aqlmPD43sQ, join: https://t.co/EKlAomDEcIArcee.ai @arcee_ai
260 Followers 192 Following Arcee is the leader in domain-adapted LLMs w/ our Small Language Models (SLMs) & our Model Merging innovations https://t.co/ZjIQuOcoknMatt Beane @mattbeane
3K Followers 219 Following Study work with intelligent machines, esp. robots. @MITSloan PhD, @Ucsb Asst Prof, @Stanford fellow, @tedtalks Book, Substack: https://t.co/OJQa0by9MTDr. Thomas Ichim @exosome
25K Followers 26K Following CEO of Immunology Based Suicide Prediction and Prevention Biotech Company Campbell Neurosciences. Working for a world without suicide.Jason Sosa @jason_sosa
1K Followers 1K Following Entrepreneur | Speaker - Founder @ Azara AI https://t.co/BDQknvIOSFCharles 🎉 Frye @charles_irl
9K Followers 2K Following ai engineer at @modal_labs. he/him. ex @full_stack_dl, @weights_biases, phd Berkeley @Redwood_Neuro.gabriel @gabchuayz
420 Followers 513 Following ✌️ Data Scientist @GovTechSG🔮 https://t.co/IiByfbW4HKChrisKraków @cmerk72
78 Followers 444 Following German expat in Krakow, ex-Google, working on AI & ML since 2000, published Graph Neural Networks in 2005Aidan @aidancramer
5K Followers 1K Following Builder & ex-CEO: Stride (acq. LiquidFriday) | PetVet (acq. HomeLifeMedia) Driving: https://t.co/yDrWOFVeB6 – Career co-pilot. Enjoying the ride.Code dropped: "Level Of Gaussians Real-Time View Synthesis for Large Scenes with Millions of Square Meters" Project: zju3dv.github.io/LoG_webpage/ Code: github.com/zju3dv/LoG
✨ I upscaled a 64x64 pixel image to 37 Megapixels 🤯 Workflow: the new Clarity AI Low Res for upscaling from 64 to 1024px and Clarity AI regular for 1024 to 6144px. I made both models open-source and am now implementing the low-resolution option to my app…
For some people, it helps to understand DSPy as a programming language for RL over natural language. But there are two RL problems: (1) learning to prompt/finetune the LMs in the DSPy program & (2) learning to act as the LM in the program. The policy for (1) isn't always a DNN.
Replacing Judges with Juries Evaluating LLM Generations with a Panel of Diverse Models As Large Language Models (LLMs) have become more advanced, they have outpaced our abilities to accurately evaluate their quality. Not only is finding data to adequately probe
GGML Flash Attention support merged into llama.cpp github.com/ggerganov/llam…
ScrapeGraphAI: A Web Scraping Python Library that Uses LLMs to Create Scraping Pipelines for Websites, Documents, and XML Files Quick read: marktechpost.com/2024/04/30/scr… Github: github.com/VinciGit00/Scr… Colab Notebook: colab.research.google.com/drive/1sEZBonB… @LangChainAI #artificalintelligence
A wealth of results that many people will find valuable. GPT-3.5 vs GPT-4 with & without DSPy optimizers. Optimized GPT-3.5 calls are consistently more accurate here than uncompiled GPT-4 calls.
Authors @AugustinToma @RonaldXie1 @SPalayew @BoWang87 won all three sub-tasks of MEDIQA-CORR 2024 with their programs. Check out their thread (above) and paper (below). The image shows the added gains by optimizing the language programs in DSPy. arxiv.org/pdf/2404.14544
SAGS Structure-Aware 3D Gaussian Splatting Following the advent of NeRFs, 3D Gaussian Splatting (3D-GS) has paved the way to real-time neural rendering overcoming the computational burden of volumetric methods. Following the pioneering work of 3D-GS, several methods have
Meta announces Better & Faster Large Language Models via Multi-token Prediction Large language models such as GPT and Llama are trained with a next-token prediction loss. In this work, we suggest that training language models to predict multiple future tokens at
If your model is weak, your paper might end up getting more citations because people are always happy to include your model as a baseline.
Let's learn how to build a financial agent that not only answers pointed questions over documents, but can perform complex computations on top of the tabular data within these docs 🧮 This requires the following layers on top of RAG: - a memory layer to track state (extracted…
Performing Complex Financial Calculations with Agentic RAG 🧮📈 Let’s build a financial assistant that can calculate percentage evolution, compound annual growth rate (CAGR), and P/E ratios over unstructured financial reports, without any human data transformation! This post by…
LlamaParse 🤝 Bedrock Knowledge Base If you're an AI developer in the AWS/Bedrock ecosystem, you can still take advantage of the advanced parsing capabilities that LlamaParse provides to build advanced RAG over complex PDFs, and we have a full tutorial on doing so ✅…
A Reference Architecture for Advanced RAG with @llama_index and Bedrock 📖 If you’re looking to build advanced RAG in the AWS ecosystem, our code repo gives you the perfect reference material for both advanced parsing and agentic reasoning, while using a full suite of AWS…
Learn how to build agentic RAG with semantic caching to speed up common queries! In this collaboration with @Redisinc, @tchutch94 and @seldo explain how to build world-class RAG applications optimized for quality, efficiency and cost. youtube.com/watch?v=mTNiGf…
Random thought, but curious if there’s room for semi-local inference (not on device, but for example, a wifi router that can run powerful LLMs). Speed, cost, privacy benefits over using APIs to power and control smart home/office devices, etc.
Performing Complex Financial Calculations with Agentic RAG 🧮📈 Let’s build a financial assistant that can calculate percentage evolution, compound annual growth rate (CAGR), and P/E ratios over unstructured financial reports, without any human data transformation! This post by…
Oh, look, I hear there's a DSPy meetup in San Francisco on Wednesday (May 1st) from 5:30 PM - 8:30 PM with @ArizePhoenix, @cohere, @weaviate_io. See you there, everyone!
We've aggregated all of the DSPy resources from the Weaviate team on one page! It is broken into two categories: 1. Hands on Learning and 2. Read and Listen 📚 DSPy round-up: weaviate.io/developers/wea… We're also very excited to host our in-person meetup with DSPy, @arizeai, and…
𝐒𝐭𝐫𝐞𝐚𝐦𝐌𝐮𝐥𝐭𝐢𝐃𝐢𝐟𝐟𝐮𝐬𝐢𝐨𝐧 - Real-time semantic drawing app is available at @huggingface Spaces! 🔥 🔥 Enjoy: huggingface.co/spaces/ironjr/…
Kangaroo Lossless Self-Speculative Decoding via Double Early Exiting Speculative decoding has demonstrated its effectiveness in accelerating the inference of large language models while maintaining a consistent sampling distribution. However, the conventional
#OnThisDay in 1993, CERN put the World Wide Web in the public domain, later made available with an open licence, allowing the web to flourish. Here we see Sir Tim Berners-Lee's proposal for the World Wide Web, revised back in May 1990. The World Wide Web will be one of the…