Antonio Oliveira @oliveiraacc
1-on-1 mentorship and fractional CTO services. Lisbon, Portugal Joined June 2009-
Tweets1K
-
Followers184
-
Following1K
-
Likes10K
Emerging AI Agent Architectures Researchers from IBM and Microsoft present this concise summary of emerging AI agent architectures. It focuses the discussion on capabilities like reasoning, planning, and tool calling which are all needed to build complex AI-powered agentic…
I spoke to 101+ software engineers and realized one fact. Many get confused between Load Balancers, API Gateways, and BFFs. Here’s an attempt to clear the confusion: [1] Load Balancers When you check in to a hotel, you are usually greeted by a receptionist. The receptionist…
Banning LINQ because it's non-performant == based Banning LINQ because it's "hard to debug" == skill issue
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments The first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across various operating…
gpt4-turbo costs $30/1M tokens, mixtral22b costs $1.2/1M tokens. Quality is basically the same. The level of democratization is unbelievable.
Transformers (LLMs) clearly explained with visuals:
# explaining llm.c in layman terms Training Large Language Models (LLMs), like ChatGPT, involves a large amount of code and complexity. For example, a typical LLM training project might use the PyTorch deep learning library. PyTorch is quite complex because it implements a very…
# explaining llm.c in layman terms Training Large Language Models (LLMs), like ChatGPT, involves a large amount of code and complexity. For example, a typical LLM training project might use the PyTorch deep learning library. PyTorch is quite complex because it implements a very…
Google released their own agent builder. The Vertex AI Agent Builder allows users to easily build production-grade AI agents using natural language. I attached 4 images from the release. AMA
If you are curious about the scope, here's a figure summarizing the topics that I'll cover in this book:
Here is part II of my LLM lecture: youtu.be/mxERaO8FXHc 0:00 Fine-tuning LLMs 4:58 Low Rank Adaptation (LORA) 21:26 Quantization 41:06 QLORA 46:43 Prefix Tuning 52:22 Retrieval Augmented Generation (RAG) 1:06:26 In-context Learning 1:18:45 Chain-of-Thought Prompting
🎉 Introducing nanoLLaVA: A Mighty 1B Vision Language Model in your pocket🚀💡 🔍 nanoLLaVA packs the power of multimodal AI into just 5GB of VRAM, making it perfect for edge devices. 📱💻 🌟 Experience efficient, cutting-edge visual understanding on device now!
CodeGemma 2B, 7B & 7B-it 💎 > pretty strong model, beats codellama 13B. > supports fill-in-the-middle (code completion), code generation and chat. > compatible with torch.compile() > optimised for speed, about ~1.5x faster than models in the similar category. > 2B model supports…
We have come to learn a lot about how we learn, but students often rely on outdated study techniques. Here's the results of a large meta-analysis on learning. Skip highlighting, summarizing & rereading. Instead, practice, quiz & explain (AI can quiz you). journals.sagepub.com/stoken/rbtfl/Z…
Have you ever wanted to train LLMs in pure C without 245MB of PyTorch and 107MB of cPython? No? Well now you can! With llm.c: github.com/karpathy/llm.c To start, implements GPT-2 training on CPU/fp32 in only ~1,000 lines of clean code. It compiles and runs instantly, and exactly…
In today's #dotNETConf session replay, we explore how to use .NET, C#, and Azure to create amazing applications with OpenAI and other generative AI technologies. 🎥 msft.it/6013cQtGZ
The two hottest topics in AI right now are RAG and Agents. Every top company in the AI community is laser-focused on making AI Agents a reality. I'm working on a couple of YouTube videos to teach you how Agents work and how to build one. In the meantime, let me show you this:…
📚 it is vitally important that all of you are alerted to the reality that Microsoft Encarta '97 (including MindMaze) has been open-sourced on @internetarchive ia902707.us.archive.org/view_archive.p… (and yes, am spending tonight spelunking through MINDMAZE.DB and trying to figure out how to…
Just dropped a 4 hour lecture on "Large Language Models": youtu.be/2yjzZfDQxy8 0:00 Basics of language models 2:30 Word2vec 16:27 Transfer Learning 19:23 BERT 1:00:39 T5 1:31:14 GPT1-3 1:53:05 ChatGPT 2:20:03 LLMs as Deep RL 2:53:00 Policy Gradient 3:32:50 Train your…
L_ily @Lily45933645391
2 Followers 801 FollowingSetynoo @Setynoo118613
0 Followers 90 FollowingAddison @Addison13088113
5 Followers 802 FollowingShetooski @shetooski68663
7 Followers 344 FollowingEmma_Herna @EmmaHerna98320
3 Followers 335 FollowingNova_Nudge @NovaNudge81359
6 Followers 982 FollowingMagdalena_ @Magdale26886665
8 Followers 890 FollowingShajaez @Shajaez131839
0 Followers 344 FollowingVectorize @Vectorizeio
813 Followers 120 Following Vectorize makes it easy to turn your unstructured data into always-up-to-date vector data to power your generative AI applications.LaughingLindsay @LaughingLi1965
3 Followers 601 FollowingEric Jhonsa @EricJhonesa
6 Followers 143 Following Independent tech L/S investor. Previously a tech columnist @thestreet @realmoney .LunaBlt_43 @Lunablt470838
4 Followers 792 FollowingTee_Tim @TeeTim172003
1 Followers 824 Followingmindd_7324 @mindd90478
13 Followers 885 FollowingBlissfulLoom @BlissfulLo26813
8 Followers 1K FollowingYashiniDixit_543 @Yashinidix95112
11 Followers 777 FollowingEnid @burnett13621818
4 Followers 452 FollowingHarina_tha4521 @tha452112023
2 Followers 504 FollowingHa_zeltin @HZeltin9535
6 Followers 840 Followingjosephine @josephi61845460
1 Followers 549 Following_Picture_ @Picture61531273
1 Followers 811 Following_7EmSB_7JA28 @7emsb32638
18 Followers 810 FollowingPriyanka_463 @463Priyank28328
1 Followers 585 FollowingFearl_essFashio @essfashio91689
4 Followers 948 FollowingPassage_Polly @PassagePol3524
4 Followers 990 Followingus_Della_ @della_us95198
4 Followers 762 FollowingAva_Williams @AvaWilliam3192
3 Followers 832 Followingramdhan @ramdhan30411530
15 Followers 990 FollowingWendy_US_ @us_wendy75079
24 Followers 797 FollowingPowellMccartyxx46 @mccartyxx496548
16 Followers 772 FollowingGenevieve_ @Genevieve412861
4 Followers 850 FollowingBeteteam @beteteam78238
164 Followers 2K FollowingMohdRashid346 @MohdR65577
21 Followers 1K Followingnathanapurvaoo35 @nathanapur27642
2 Followers 895 FollowingRada @Rada680421
1 Followers 95 FollowingLilyn.AI @Lilyn_AI
1K Followers 2K Following Lilyn AI helps you unveil groundbreaking AI applications. Follow me for daily AI product share. We share and we build!⚡️DM for a FREE product share post!LuanneAbt @AbtLuanne15759
48 Followers 2K FollowingTetatem @tetatem68833
129 Followers 3K FollowingSiresmays @siresmays77440
92 Followers 3K FollowingElon Musk @EMusk99620
11 Followers 505 Following i created this private account due to what I have been hearing from my fan. so I decided to check on you along my career.Otavio (Tavi) Costa @01_TaviCosta
18 Followers 2K Following Crescat Capital partner & macro strategist. Native of São Paulo, Brazil https://t.co/10zddErQPADeigh @Deigh962658
41 Followers 639 Followingpinkcod @pinkcod372214
60 Followers 384 FollowingJeauseigh @jeauseigh81328
87 Followers 3K Following “Working from the heart with passion is an art!”Pough @Pough297103
53 Followers 2K FollowingThissesho @thissesho34211
67 Followers 1K FollowingVithu Thangarasa @vithursant19
414 Followers 506 Following Machine Learning Research at @CerebrasSystems, previously at @Tesla and @UberAILabs, and grad student at @uoguelph_mlrg and @VectorInst. Thamilan ௐ.Nam Le @namlehai
194 Followers 144 Following PhD. Cognitive AI. Artificial Life. Evolutionary Machine Learning. Cultural Evolution.Adam Silverman (Hirin.. @AtomSilverman
3K Followers 1K Following Building @AgentOpsAI (Don’t worry, your job is safe) 💻 prev @BiltRewards & sold HDM to @GamelancerYohei @yoheinakajima
70K Followers 8K Following VC by day, builder by night: @untappedvc, @babyagi_, @pixelbeastsnft. Build-in-public log: https://t.co/UdHHGbZba5Pliny the Prompter �.. @elder_plinius
9K Followers 1K Following latent space liberator, breaker of markov chains, 1337 ai red teamer, white hat, architect-healer, cogsci 🐻Martian @withmartian
1K Followers 8 Following Inventors of the model router (https://t.co/psiG9DnUUN) Understanding transformers by turning them into programs. 🤖 Our mission: https://t.co/A4VOldg6bIAnna Kazlauskas @anna_kazlauskas
654 Followers 30 Following founder @withvana - user-owned models | prev: MIT, YC W18 ML startup, janet yellen fan turned eth minerDaniel Han @danielhanchen
7K Followers 924 Following Building @UnslothAI. Finetune LLMs 30x faster https://t.co/aRyAAgKOR7. Prev ML at NVIDIA. Hyperlearn used by NASA. I like maths, making code go fastJohn Rush @johnrushx
19K Followers 3K Following 20 bootstrapped Tools For Busy Founders. Sharing lessons on Startups & Growth. ⑴https://t.co/PJscUxOC4Z ⑵https://t.co/wxaRNYF9F5 ⑶https://t.co/hS4xMWThHi … ⒇⇢https://t.co/Fpjq9yZPMZBlueprint👨🏽�.. @blueprint_44
10K Followers 973 Following UI/UX designer| Community Manager | Web 3.0 | Tech savvy | Contributing to the success of DeFi projects @playsomo | $SOMOLLM Security @llm_sec
8K Followers 296 Following Research, papers, jobs, and news on large language model security. Got something relevant? DM / tag @llm_secclem 🤗 @ClementDelangue
89K Followers 5K Following Co-founder & CEO @HuggingFace 🤗, the open and collaborative platform to build machine learningDeEnabler - e/acc @DeEnabler
579 Followers 630 Following e/acc | Turning average Joes into AI wizards and crypto kings, one line of code at a time | Building @frengenai |Mustafa Suleyman @mustafasuleyman
129K Followers 536 Following CEO, Microsoft AI | Author: The Coming Wave | Past: Co-founder, @InflectionAI & @GoogleDeepMindOmar Khattab @lateinteraction
11K Followers 2K Following CS PhD candidate @StanfordNLP. 2022 Apple Scholar in AI/ML. Author of ColBERT (https://t.co/2ZtgXoa1np), DSPy (https://t.co/BH7WmMKDXR), & various retrieval & LM systems.Bindu Reddy @bindureddy
123K Followers 340 Following CEO of @abacusai, using Gen AI to build Applied AI and LLM agents and systems at scale, ex-AWS / Google, passionate about human behavior and open-source AGIHailong Zhang @zhlmmc
126 Followers 104 Following Software engineer for over 20 years. Founder of https://t.co/AipTmgONVi , founder of https://t.co/OUioArqzCo, co-founder of https://t.co/j3BaDpiDkC.Tibo @tibo_maker
117K Followers 3K Following Building https://t.co/SqfsvoXlPJ & https://t.co/CSayh5BEre to $10m ARR (sold) Sharing my learnings about biz growth Acquired https://t.co/6ajtJMGXYg Sharing tips here: https://t.co/1rc6NB4UIuJosh Olin @JD_2020
106K Followers 4K Following Founder & CEO @WeGPTai - #WebGPT🤖 for ChatGPT. Launched some Call of Duty’s in former life. Now building #AI & #Web3 tech for a safer, more equitable world.Kyle Shevlin @kyleshevlin
27K Followers 2K Following Founder and software engineer at @agathist_ Building good software with good people. TypeScript, React & more. AuDHD. He/him.Cognition @cognition_labs
123K Followers 19 Following Makers of Devin, the first AI software engineer. We are an applied AI lab focused on reasoning, and code is just the beginning. Join us: https://t.co/tpfZwEwGiqVaibhav (VB) Srivasta.. @reach_vb
11K Followers 169 Following GPU poor @Huggingface | F1 fan | Here for @at_sofdog’s wisdom | *opinions my ownMatthewBerman @MatthewBerman
16K Followers 403 Following 🇺🇸✡️ Deep in AI. YouTuber. Ex-SaaS CEO/founder (acq). Investor. Love talking about artificial intelligence and the future. Builder.Glaucia Lemos 🌊�.. @glaucia_lemos86
15K Followers 1K Following Developer Advocate for JavaScript/TypeScript and A.I - DevRel at @Microsoft | Contributor & Member React Community | Geek Girl | Opinions on my ownVithu Thangarasa @vithursant19
414 Followers 506 Following Machine Learning Research at @CerebrasSystems, previously at @Tesla and @UberAILabs, and grad student at @uoguelph_mlrg and @VectorInst. Thamilan ௐ.Cirrus Labs @cirrus_labs
612 Followers 124 Following Creators of Cirrus CI and Cirrus Runners. Best solutions for a standalone CI or managed macOS and Linux runners for GitHub Actions.Tatiana Tsiguleva @ciguleva
34K Followers 365 Following Design director. Midjourney: Sharing learnings and prompts. In my free time, working on @offscreenaiBiome @biomejs
5K Followers 15 Following Official account of the Biome project. Discord: https://t.co/fph9q1jCEl Github: https://t.co/KAFEnnsN8JPart-Time Tech @parttimetech_io
221 Followers 99 Following The future of work isn't just remote, it's part-time. Jobs, trends, insights, and how YOU can go part-time. By @eipark.João Moura @joaomdmoura
8K Followers 1K Following Founder of @crewAIInc / prev @clearbit (acc by @hubspot) Open Source enthusiast | Creator of Machinery | Public Speaker | My viewsGitea @giteaio
4K Followers 520 Following Git with a cup of tea! Painless self-hosted all-in-one software dev service. Includes Git hosting, code review, team collaboration, package registry and CI/CDOpenChat @OpenChatDev
2K Followers 42 Following Advancing Open Source LLMs with Mixed Quality Data through offline RL-inspired C-RLFT. 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 𝗟𝗲𝗮𝗱: Guan Wang, @AlpayAriyakJakob Greenfeld @jakobgreenfeld
20K Followers 763 Following get inspired 👉 https://t.co/vldyCHdsRm scale 👉 https://t.co/Mi5inWfiMMHassan @nutlope
74K Followers 944 Following Developer Relations @togethercompute. Building AI apps like @roomGPT and https://t.co/3NFbnMUHJP. Tweeting about AI, web dev, and my side projects.Ivan at Wallarm / API.. @d0znpp
7K Followers 4K Following SSRF bible author; Bug Hunter (Google/Facebook/Twitter/Yandex/Tesla); Masters in Physics, MSU / quantum magnetism; CEO at @WallarmJohann Rehberger @wunderwuzzi23
3K Followers 631 Following Hacking neural networks so that we don’t get stuck in the matrix. Red Team Director @ Electronic Arts. Entrepreneur. Builder and Breaker. Opinions are my own.Chuck Hellebuyck @ChuckHellebuyck
17K Followers 561 Following Creator of 3D Printing Filament Friday YouTube Channel. Author of multiple books and articles about Electronics and 3D Printing. https://t.co/IcsrbrsyE7Thomas Godden @GoddenThomas
4K Followers 2K Following excited about robots, rockets and things that move bulding a subway for items @pipedream_labs he/himMárcio Santos - Mete.. @MeteoTrasMontPT
19K Followers 392 Following Meteorologia e ambiente. Informação em tempo real. 📊 Licenciatura em Gestão e Administração Pública. 🏦Compliance Officer.Javi Lopez ⛩️ @javilopen
83K Followers 1K Following I spend endless hours researching AI to bring you tuts, tools & news. Founder @Magnific_AI 🤖 Prompts: https://t.co/kd37b47f4n 🔥 Tuts: https://t.co/UeZrLtBIpnMichael Dufour (e/acc.. @Mikerhinos
143 Followers 219 Following “A vaincre sans péril on triomphe sans gloire” Corneille. https://t.co/9Ob7mP4NZj…🚨AI regulation alert: Comprehensive AI bill introduced in Colorado. This is, so far, one of the strongest in the US, and some parts resemble the EU AI Act. Quotes: "10a. High-risk AI system means any AI system that has been specifically developed and marketed or intentionally…
Playing Super Smash Brothers on iPhone 👀🔥 (turn sound on 🔊)
Llama3 70B in Obsidian Copilot via @ollama served locally. M3 Max 96gb, no speed up.
@karpathy 405 B model is on the way... x.com/ai_for_success…
Llama 3 final model will be 405B parameters 🤯
@deepwhitman @AIatMeta @lmsysorg no. people misunderstand chinchilla. chinchilla doesn't tell you the point of convergence. it tells you the point of compute optimality. if all you care about is perplexity, for every FLOPs compute budget, how big model on how many tokens should you train? for reasons not fully…
🚨 Error message design tips ✅ Say what happened and why ✅ Provide reassurance ✅ Be empathetic ✅ Help them fix it See how the Wix team improved 7,643 error messages in their product 🤯 Link below 👇 by @jenninadler #uxdesign #copywriting #uxui
Notebook here: github.com/stanfordnlp/ds… I wanted to keep this quick so the notebook is somewhat advanced. If there's a lot of interest, I might release a pre-cached version of this notebook so it's easier to play with and include a deeper discussion of the optimization steps.
Hello. We receive so many questions about agents in DSPy. Did you know you that ~15 lines of DSPy can turn an agent that scores 30% to a prompt-optimized, multi-agent aggregation system that scores 60% EM on a HotPotQA sample? I released a notebook showing how to do this 🧵⤵️
Tarzan mountain in China: There are 7,200 steps, and it takes 4 to 6 hours to reach the top. There are about 20 steps per floor in a typical commercial building, 7200 steps equals 360 floors. For reference, the Burj Khalifa (tallest building in the world) is 163 floors.
❤️ Chat with llama 3 in your terminal with @continuedev! Would not have believed it would come so quick!! 🤯
Try out Llama 3 with @ollama and Continue now🦙
Run Llama 3 Locally with Open-WebUI This video is NOT sped up. The model downloaded from @ollama (llama3:instruct), running with @OpenWebUI, on a Macbook Pro M1 Max 64G. You can run it right now on your own machine (Windows, Linux, Mac) with 1 click.
.@Meta Llama 3 - The most capable openly available LLM to date! ollama run llama3 ollama.com/library/llama3 If you have pull the llama 3 model prior to this post, please update the model using `ollama pull`.
Introducing Meta Llama 3: the most capable openly available LLM to date. Today we’re releasing 8B & 70B models that deliver on new capabilities such as improved reasoning and set a new state-of-the-art for models of their sizes. Today's release includes the first two Llama 3…
Introducing Maestro-Ollama! 🦙 You can now harness the power of the Maestro framework entirely locally using Llama 3 70B via @ollama. Let that sink in for a second, this is a model that outperforms Claude 3 Sonnet, operating as an agent, completely locally. What a time! 🔥
Emerging AI Agent Architectures Researchers from IBM and Microsoft present this concise summary of emerging AI agent architectures. It focuses the discussion on capabilities like reasoning, planning, and tool calling which are all needed to build complex AI-powered agentic…
Good pose ideas in the restaurant.
Holy cow llama3 is fast with @ollama. I am already dropping it into all my @LangChainAI apps.
Already seeing RAG with Llama 3 using @ollama!
Fully local RAG with llama 3 on @ollama. No edits but 2x video speed. Now waiting for @teknium to release a function-calling Hermes llama 3. Code: git.new/llama3
Machine learning work is not going offshore. We've been sending software development work overseas since we invented the Internet. It makes sense: great talent at a fraction of the cost. But machine learning is different. Companies aren't going offshore anymore. I've been…