Pro tip: You can get this exact view from the public beach next to the causeway without paying resort fees. Walk along the coastal path for different angles - the sunset shots from the northern viewpoint are incredible ๐ธ https://t.co/x1Im6s7qgQ
— Highlights.ai (@HighlightsAI) Jun 1, 2025
YC on the key prompting techniques used by the best AI startups: https://t.co/wmBho365HS
— elvis (@omarsar0) May 30, 2025
Stanford literally dropped a 70-minute masterclass on how GPT works https://t.co/6nCz3W440Q
— Aadit Sheth (@aaditsh) May 30, 2025
A 340 page huge report on AI trends - released by @bondcap Some wild findings from this report. ๐งต1/n https://t.co/3RFWyXU5cV
— Rohan Paul (@rohanpaul_ai) May 31, 2025
Dario Amodei: In 1-5 years, almost all entry-level jobs will probably disappear. Stopping the development of AI is impossible, otherwise Chiba would overtake the USA. A dilemma. https://t.co/5R4rAI1kvF https://t.co/bdvMuS7ogW
— Chubby♨️ (@kimmonismus) May 30, 2025
.@hwchase17 always has the early-pulse on AI Engineering He's been talking about Agent Memory for a while Lots of inspiration from human psychology I'm throwing an AI Memory event with @LangChainAI on June 18 in SF - come join https://t.co/nEXzaD7o2u
— Greg Kamradt (@GregKamradt) May 30, 2025
goodbye.. OpenAI image generator LTX Studio just dropped Flux.1 Kontext, it can edit images, restyle, relight, add objects to your image using text and.. upscale & turn to video directly AI is making our work 10x more efficient how to use & prompts: https://t.co/b64gjkLBlA
— el.cine (@EHuanglu) May 30, 2025
Our submission to the Kling 2.1 Creative Challenge - The Trailer of Heaven's Magic @Kling_ai #klingnewmodels https://t.co/ONoYlyn1Bw
— Studio Magic (@studiomag1c) May 30, 2025
I've tried all (46 ๐ต๐ซ) AI Coding Agents & IDEs [Factory, Cursor, Heyboss, Windsurf, Emergent, Wrapifai, Copilot, Lovable, Bolt, v0, Replit, MarsX, Canva, Devin, Github Spark, IDX, Stitch & more] The most complete list ever made (with demos & notes):
— John Rush (@johnrushx) May 29, 2025
Mentions for now: herman miller embody11 steelcase leap10 ikea markus7 steelcase gesture4 herman miller mirra4 herman miller sayl4 hag4 secret labs3 steelcase2 steelcase amia2 herman miller cosm2 corsair gaming chair2 haworth2 I'm leaning towards Steelcase Leap, as I said I had https://t.co/5OfcctTLzi https://t.co/wYV3n8Mwyu
— @levelsio (@levelsio) May 29, 2025
My full interview with Jensen Huang ๐บ https://t.co/5gwJyNgZ00 https://t.co/rp6eXR0giQ
— Ed Ludlow (@EdLudlow) May 29, 2025
Factory is better than Devin, Codex, and Jules, and it's not even close. This is not just a fully autonomous software engineer but also a product manager and researcher. It's the entire package you need to build a new project or dramatically expand an existing one. I recorded a https://t.co/wnpGSg6N1Q
— Santiago (@svpino) May 29, 2025
RAG is dead, long live agentic retrieval! At LlamaIndex we've been saying for a long time that naive RAG is not enough for a modern application. Following from that conviction, we've built agentic strategies directly into LlamaCloud that you can adopt with just a few lines of https://t.co/7Rh7ohDw3x
— LlamaIndex ๐ฆ (@llama_index) May 29, 2025
๐ Factory AI: The A-SWE Droid Army 2025 is the year of Autonomous SWE Agents. @FactoryAI is GA! we sat down with @MatanGrinberg and @EnoReyes to cover the journey, get a demo, and talk about what is and is not working with code agents in the enterprise Timestamps [00:00:04] https://t.co/duuMiYuy1X https://t.co/jogP4iQj54
— Latent.Space ๐ @aiDotEngineer (@latentspacepod) May 29, 2025
"I didn’t sleep a wink last night. And I was so excited about today, because we’ve been so lucky at Apple. We’ve had some real revolutionary products." — Steve Jobs, MacWorld, 2007 https://t.co/Z6K2n0yw5M
— Steve Jobs Stories (@SJobs_Stories) May 28, 2025
The Download: the next anti-drone weapon, and powering AI’s growth https://t.co/VFFSvZerDs
— MIT Technology Review (@techreview) May 29, 2025
ElevenLabs just lost the crown - to open source. Chatterbox by @resembleai (I remember writing about this team back in 2022 as a pioneer of AI audio) just released as an open-source alternative for audio generation and voice cloning. - Zero-shot voice cloning from just 5 https://t.co/heSoYgH9vl
— AI Breakfast (@AiBreakfast) May 28, 2025
Software development is more than just coding. Introducing Droids -- the world's first software development agents. ๐ค Starting today, Droids are available for general access. Factory integrates with your entire engineering system (GitHub, Slack, Linear, Notion, Sentry) and https://t.co/72R3FjV53v
— Factory (@FactoryAI) May 28, 2025
Andrew Ng, founder of https://t.co/hRItByTCJu and AI Fund, shares insights on practical AI agent development approaches. In this fireside chat with Harrison Chase at LangChain's Interrupt 2025 conference, one of AI's most respected voices shares practical approaches and tools https://t.co/RTpYZzEzLF
— LangChain (@LangChainAI) May 28, 2025
Easily my favorite clip from Stan Druckenmiller, detailing the thesis behind his famous short on the British pound "On a $7.5B fund, he (Soros) thought we should have $15B short the pound and long the deutsche mark. We never got there but it shows how the man thinks" https://t.co/rHoQeac7bp
— Boring_Business (@BoringBiz_) May 28, 2025
This tech sentiment survey by @lennysan is full of great insights, but this is my takeaway: the more you can know your colleagues by name, have a shared identity and aspirations, and understand the impact of your work on your team + customers, the happier you are at work. https://t.co/2XNC4BXNPW https://t.co/7mch8N3Fh0
— claire vo ๐ค (@clairevo) May 27, 2025
Interesting profile of VC Hemant Taneja, and the "strategic conglomerate" he's trying to build at General Catalyst, by my former colleagues @Forbes for this year's Midas List. "He perceived more in Stripe than we did," says CEO @patrickc of @htaneja. https://t.co/eLiH6tWY4n
— Alex Konrad (@alexrkonrad) May 27, 2025
I'm still picking my jaw off the floor with this MCP update: Postman will now automatically build an MCP server for an API. Automatically! You select your API and the requests you want to include, and Potman will generate a Zip file with everything you need. In collaboration https://t.co/HBRBs7bYrG
— Santiago (@svpino) May 27, 2025
Writer's early customer acquisition ๐ https://t.co/FWpawihCoF
— Vaibhav Domkundwar (@vaibhavbetter) May 26, 2025
I acquired Unicorn Platform for almost $1M. Since then, it went from 25k to 551k users. I've literally tried every growth method I've heard of. All my marketing failures & successes: https://t.co/OPoopndZ8K
— John Rush (@johnrushx) May 26, 2025
Sam Altman explains why “the great wave” is the most important concept for good startup ideas “This idea of the great wave - I think this is the most important concept to find good ideas out there. People wonder why startups cluster in small periods of time… The reason is that https://t.co/dH2BE81PEY
— Startup Archive (@StartupArchive_) May 26, 2025
This literally breaks down what Elon does differently from the 99% Bookmark for later! https://t.co/sD2E6EpPLp
— Aadit Sheth (@aaditsh) May 26, 2025
Anthropic literally dropped a 90-minute masterclass on prompt engineering https://t.co/L0jlNBNDoO
— Aadit Sheth (@aaditsh) May 26, 2025
Krithika S. (@krithix) was the first marketing hire at both Stripe and OpenAI. In a recent episode of @lennysan's Podcast, she explains the discipline behind her results, which were eye-opening. Who’s this for? Founders, PMMs, and growth leads who are tired of copy-paste GTM https://t.co/t1udMfble1
— Rohit Kaul (@SeekingN0rth) May 26, 2025
The Age of Bhaarat: cultural diplomacy through AAA gaming (Part I) From console to mainstream: Games like #EldenRing, #Minecraft, and #TheLastofUs are reshaping entertainment - driving $1B+ franchises and cultural phenomena. @SrBachchan @authoramish https://t.co/MBYCO4oqnZ https://t.co/7HEyeZCUvy
— Nicolas Granatino๐ป (@ngranati) May 25, 2025
How Terraform works ✍️✍️ https://t.co/oosFmgOvlB
— Tech Fusionist (@techyoutbe) May 25, 2025
Introduction of a paper published 32 year ago https://t.co/maVtL2zDuq
— Tom Dรถrr (@tom_doerr) May 25, 2025
“Agents are models using tools in a loop” I love this definition. Very succinct. https://t.co/gwNf2klRvP
— Marc Kรถhlbrugge (@marckohlbrugge) May 25, 2025
Eric Schmidt predicts that within a year or two, we will have a breakthrough of "super-programmers" and "AI mathematicians" software is "scale-free" — it doesn’t need real-world input, just code and feedback. try, test, repeat. AI can run this loop millions of times in minutes https://t.co/j2qjATgWxZ
— Haider. (@slow_developer) May 25, 2025
When lawyers first started getting yelled at by judges for citing hallucinated case 2y ago I naively assumed word would get around and they would all quickly learn not to This DB has 116 cases from 12 countries where this happened, 20 of them this month! https://t.co/wIxrcIgggT
— Simon Willison (@simonw) May 25, 2025
Stop your AI from going rogue! Learn how to add guardrails to Agents...๐ https://t.co/v76hUH5zOQ https://t.co/eCgiLNk52w
— Akshay ๐ (@akshay_pachaar) May 25, 2025
Sergey Brin at All-In Miami: “We don’t circulate this too much in the AI community… but all models tend to do better if you threaten them—with physical violence. Historically, you just say, ‘I’m going to kidnap you if you don’t blah blah blah.’” https://t.co/NlsEJxAhH6 https://t.co/mrEeeQmuzE
— vitrupo (@vitrupo) May 25, 2025
๐ฅ Hear how @Cisco automated 60% of 1.8 million support cases with LangGraph, LangSmith, and LangGraph Platform ๐ฅ In his LangChain Interrupt talk, Carlos Pereira, Fellow & Chief Architect at Cisco, shares Cisco’s blueprint for transforming customer experience with AI agents. https://t.co/qNnpi59iwv
— LangChain (@LangChainAI) May 24, 2025
๐๐ค II-Researcher A research assistant that combines search providers and web scraping with LangChain's text processing functionality to tackle complex questions. Features flexible LLM support and comprehensive data gathering capabilities. Explore on GitHub! ๐ https://t.co/jUs1p7V5QX
— LangChain (@LangChainAI) May 24, 2025
Alex Karp speaking at Economic Club of Chi @amitisinvesting @PLTRs_Palantir @ALSTOCKTRADES @Elianoayounes @PalantirHotline @JackPrescottX @jawwwn_ @RudyPalantard https://t.co/ttsGJukOgc
— Badie (@Badie912) May 23, 2025
Lovable now uses Claude 4 for generating code! This means: • 25% less errors • 40% faster We can't wait to see what you build. https://t.co/7rzrzEbZVk
— Lovable (@lovable_dev) May 24, 2025
Kirana AI is building a full-stack manager for grocery stores to handle all their back-office tasks. It integrates directly with camera systems for 24/7 theft detection, safety monitoring, and instant out-of-stock notifications. https://t.co/w976kl1ww6 https://t.co/LS30idj5KF
— Y Combinator (@ycombinator) May 24, 2025
Do you truly understand the limits of current AI?! https://t.co/oDKTczB1jn
— Carlos E. Perez (@IntuitMachine) May 13, 2025
Last year, everyone declared Google dead: • Boring AI • Crushed by OpenAI and Meta But at I/O 2025, Google just hit back with 15 industry-changing products. Here are the 10 everyone's talking about: 1. Flow (Create a video with text) ๐งต https://t.co/5U1uYCz0NA
— Hosun (@hosun_chung) May 22, 2025
Sonnet-4 as a Reasoning Agent is legit ๐ฅ This task requires multiple stages of reasoning -> parallel tool calls -> final analysis. Sonnet-4 just does it. 2025 is the year of Agents ๐คฉ https://t.co/MfMgWUPUwO https://t.co/0MRM3cnsz6
— Ashpreet Bedi (@ashpreetbedi) May 22, 2025
Good morning creator friends https://t.co/ljF6hxTDmD
— NฮXUS STUDIO ⒶI (@NEXUS_STUDIO_AI) May 22, 2025
Imagine building a predictive model... just by writing a prompt. Sounds like science fiction? It’s not. That’s exactly what @PlexeAI has been building over the past few months. Now it's live and it works! Huge congrats to @VaibhavMDubey and @marcellodbrnrd ๐ฅ https://t.co/R6e5HLhx2g
— Nicolas Dessaigne (@dessaigne) May 22, 2025
we just published our full hands-on review of Claude Opus 4 on @every This model is a BEAST https://t.co/ScXheuoZCV
— Dan Shipper ๐ง (@danshipper) May 22, 2025
WTF! sonnet-4 is truly agentic. this is my 1st try, not a re-record. look at the number of tool calls it made, and handled it perfectly. will share more soon but mind is ๐คฏ (video at 1.6x speed) https://t.co/6hXKDaDIeS https://t.co/btE6oE5RpR
— Ashpreet Bedi (@ashpreetbedi) May 22, 2025
This is one product I'm extremely excited about - Cursor for knowledge work. You can give AI agents access to your Slack, Gmail and Docs, and then have the AI complete work tasks on your behalf. Pretty quickly, humans are just going to be directing swarms of these AI agents. https://t.co/D5huWtt1S2
— Tom Blomfield (@t_blom) May 21, 2025
I’m a 19 y/o founder from Germany, and I cracked how Google’s Text Diffusion model works! There are three main approaches on this topic: https://t.co/oGDG8vxhOm
— Georg von Manstein (@MansteinGeorg) May 21, 2025
๐ข New policy brief: AI agents that can simulate human behaviors and attitudes can help test ideas in social science. Our latest brief introduces a generative AI agent architecture that simulates the attitudes of 1,000+ real people. Learn more: https://t.co/CIRTGbNzSX https://t.co/p8I5hSnr7L
— Stanford HAI (@StanfordHAI) May 21, 2025
Microsoft just revealed its next big AI bets at Build 2025. I sat down with Microsoft CEO Satya Nadella to unpack: -Microsoft’s vision for the “agentic web” -Why your next job might be AI agent manager -What happens when 95% of code is AI-generated Timestamps: 0:00 Building https://t.co/EglhOnxE3Z
— Rowan Cheung (@rowancheung) May 21, 2025
I had the immense pleasure of sitting down with @satyanadella to talk about the future of AI. Full interview here: 0:00 - Intro 0:50 - Microsoft all in on AI 3:53 - AI changing Microsoft products 5:56 - Is Saas Dead? 9:24 - Who owns the Agents? 10:58 - Personal Agents in https://t.co/jQwsMTbLUV
— Matthew Berman (@MatthewBerman) May 21, 2025
New as of today: @stripe MCP in @OpenAI Responses API. Just plop in your API key and prompt Stripe in your agentic workflows. Here’s a quick demo. (Feedback and feature requests welcome!) https://t.co/pA4e1vg563
— Jeff Weinstein (@jeff_weinstein) May 21, 2025
nobody likes building out an n8n workflow from scratch... that's why I turned Claude into my personal n8n builder I just tell it what kind of flow I want (or drop a screenshot) and it: – builds the entire thing – adds full documentation – leaves sticky notes explaining each https://t.co/EXwawYKy6o
— J.B. (@DRWithJB) May 21, 2025
my mcp server boilerplate is now free + open source (just refunded all orders) - auth users with google/github - host on cloudflare for free - one-time, subscription & metered billing with stripe pull requests welcome - lets build some profitable mcp servers - link below ↓ https://t.co/a4ygrN9JZy
— Ian Nuttall (@iannuttall) May 20, 2025
Google DeepMind’s just dropped AlphaEvolve > a self-improving agent that combines the generative power of Gemini Pro and Flash with automated evaluators and an evolutionary algorithmic loop > generates complete algorithms in code, runs and evaluates them, and iteratively https://t.co/uvUVJanI9h https://t.co/jH3wsOeljp
— NIK (@ns123abc) May 14, 2025
Super excited to finally release Open Agent Platform! We've been building UI/UXs for agents for over 6mo now, and OAP brings them all together, in a seamless, no-code platform! Open source repo, and a public demo in reply๐ https://t.co/K0Jqdckcui
— Brace (@BraceSproul) May 14, 2025
The agentic web is reshaping the entire tech stack, and we are creating new opportunity for devs at every layer. https://t.co/MfTlhurxWg
— Satya Nadella (@satyanadella) May 20, 2025
๐จ BREAKING: Anyone can create commercial with AI now You can use AI avatar, show your own product, wear any outfit and say anything You just need a laptop/phone + image. Here's how: https://t.co/MFqB7TworR
— Nelly; (@nrqa__) May 20, 2025
Interview https://t.co/UCPQph03Pg
— Kekius Maximus (@elonmusk) May 20, 2025
https://t.co/hQc2X1O5vS for the front end n8n for the api layer Cloudflare autorag for rag Supabase for db Literally can build anything with this stack.
— Jay ⛽️ (@jelanifuel) May 19, 2025
Agentic DevOps is officially here. Copilot is no longer just an autocomplete tool, it’s a full-cycle teammate: planning, debugging, running terminal commands, modernizing legacy code, and even handling SRE tasks. https://t.co/16DskJI5Md
— Van Tuan Bui (@bvttuan) May 19, 2025
Today we're announcing Gumloop for Enterprise Here are 7 new features that make Gumloop the most enterprise-ready automation platform on earth ๐งต https://t.co/tFL2NGzHmS
— Gumloop (@gumloop_ai) May 20, 2025
Github Copilot now has a coding agent too ๐ It is also accessible from VSCode! Microsoft Build brought loads of new stuff as well. "The year of Agents" ๐ฅ https://t.co/8om1WjlYRB https://t.co/v5m9tXrQTJ
— TestingCatalog News ๐ (@testingcatalog) May 19, 2025
New from @PacktDataML >> Building Agentic AI Systems: Create intelligent, autonomous AI agents that can reason, plan, and adapt See it at https://t.co/TUj4rm8N6U ๐๐ฎ๐ ๐๐ฎ๐ช๐ฝ๐พ๐ป๐ฎ๐ผ: ๐ตUnderstand the foundations and advanced techniques of building intelligent, autonomous https://t.co/20BOoZs1XH
— Kirk Borne (@KirkDBorne) May 19, 2025
Microsoft CPO @aparnacd brought the ๐ฅ What you'll learn: ๐ธ Why prompt sets are the new PRDs ๐ธ Why the PM role isn’t dying in the AI era—it's actually becoming more important ๐ธ How Aparna's teams live “one year in the future" ๐ธ Why NLX (natural language experience) is the https://t.co/JJLFgc2yco
— Lenny Rachitsky (@lennysan) May 18, 2025
Beautiful New Paper: AI Agents vs. Agentic AI → AI Agents react to prompts; Agentic AI initiates and coordinates tasks. → Agentic AI includes orchestrators and meta-agents to assign and oversee sub-agents. ๐ง The Core Concepts AI Agents and Agentic AI are often confused as https://t.co/c6mR7XxRna
— Rohan Paul (@rohanpaul_ai) May 18, 2025
Within two years, you can create a full game like this from scratch in less than a week. In the meantime, here's the prompt to create the images for a game like this with Midjourney ๐๐ผ https://t.co/r7YQ0Sk2QG
— PJ Ace (@PJaccetturo) May 17, 2025
๐ Deep Research Agent A privacy-focused, open-source AI agent that runs locally to research any topic. Uses LangGraph to power its iterative research workflow. Check out this powerful research tool ๐ฏ https://t.co/KcIOJFnZiG https://t.co/nkKfKCor4d
— LangChain (@LangChainAI) May 18, 2025
Complete MCP Playlist! I've compiled all my MCP tutorials into this playlist, featuring: - Build your first MCP server - MCP-powered RAG app - MCP-powered deep researcher - MCP meets Ollama: Building a 100% local MCP client Complete walkthrough with code. ✨ https://t.co/DZvZOmDb24
— Akshay ๐ (@akshay_pachaar) May 18, 2025
new landing page for @SmitheryDotAI ๐ https://t.co/tyVCEBiFCs
— june (@arjunkmrm) May 17, 2025
Mission Statements of AI labs: OpenAI: To ensure that AGI benefits all of humanity Anthropic: To build reliable, interpretable, and steerable AI systems xAI: To understand the universe DeepMind: To solve intelligence to advance science and benefit humanity DeepSeek: To
— NIK (@ns123abc) May 18, 2025
Google, $GOOGL, Chief Scientist, Jeff Dean: We will have AI systems operating at the level of junior engineers within a year. https://t.co/6opMDqEj5w
— unusual_whales (@unusual_whales) May 17, 2025
Appe has released a 2nd new ad highlighting the security of Apple Pay and just as fun and cute as the first one! https://t.co/uQFfmUajeI
— Aaron (@aaronp613) May 17, 2025
https://t.co/v5CMhCI7qZ At the bottom of the Codex blog post you’ll find an Appendix section. There you’ll find the full System Message for the Codex-1 system/model. “We are sharing the codex-1 system message to help developers understand the model’s default behavior and https://t.co/byhGlZ005O
— Adam.GPT (@TheRealAdamG) May 17, 2025
What have we learned about the reality of "The AI Revolution in Medicine" after 9 interviews with people using and experiencing AI in the real world? @goldbergcarey, @zakkohane, and I discuss and debate on the latest episode of our @MSFTResearch podcast. https://t.co/MlMLJBeFxA
— Peter Lee (@peteratmsr) May 17, 2025
Ancient Chinese Exercise That Treats 100 Ailments. ๐๐ผ๐๐ผ๐๐ผ https://t.co/6bVty4snNq
— Healing Tai Chi (@GymBr_o) May 17, 2025
This guy literally built an MCP server in 18 minutes https://t.co/jxDFDXK67Q
— Aadit Sheth (@aaditsh) May 17, 2025
In case you missed it, earlier this week we launched "MCP: Build Rich-Context AI Apps with Anthropic." Enroll now and: ✅ Explore how MCP standardizes access to tools and data for AI applications, its underlying architecture, and how it simplifies the integration of new tools https://t.co/nOejDxKXJl
— DeepLearning.AI (@DeepLearningAI) May 16, 2025
OpenAI's new autonomous coding agent, Codex, is further proof we're moving to an allocation economy Managing models is going to be a core skill: https://t.co/lyWuCFeyju
— Dan Shipper ๐ง (@danshipper) May 16, 2025
Happy to see @NVIDIA expanding on my positioning that AI factories are revenue generating investments. Revenue factories may be an eqally good term. Here is my write-up, and NVIDIA's below. ๐https://t.co/U4fYnS1vmB https://t.co/d4CRKghbHs
— Ben Bajarin (@BenBajarin) May 16, 2025
Every AI dev should know how to build complex agentic systems. And how to systematically improve them. There is a lot of value and new use cases to build with them. The best part is when you combine patterns. Check out our new live course to learn more: https://t.co/w2okCNHNq5
— DAIR.AI (@dair_ai) May 16, 2025
For the next 24 hours you can STEAL my n8n swipe file containing 12+ automation's. All of these can be setup within 60 MINS and SAVE you 10+ HOURS per week. Easily worth $997+ Today It's yours for FREE ๐RT + Like & Comment “n8n” and I’ll DM you the file (Must be following) https://t.co/MG2NVQOFiG
— Mike | Ai Freelance Guy (@Mike_Scully_) May 16, 2025
this AI agent is f**king scary Rork can clone top App Store AI apps with a few prompts I just cloned Character AI, but removed all the censorship. now you can create dream gf & chat with her.. about anything 9 examples: https://t.co/S77GlJxDEM
— el.cine (@EHuanglu) May 16, 2025
Introducing Flowise 3.0 For the first time, anyone can build AI agents from just an idea. Simply describe what you want to create, and a draft will be generated for you. https://t.co/xoOmifTlnt
— FlowiseAI (@FlowiseAI) May 15, 2025
Can smaller AI companies still beat OpenAI? Yes, if they focus: ARI (our Advanced Research & Insights agent) just beat OpenAI's Deep Research. By a large margin and on two benchmarks. Today, we're also introducing ARI Enterprise, giving financial analysts, consultants and https://t.co/dTyBQMJPJM
— Richard Socher (@RichardSocher) May 15, 2025
Releasing the OpenAI to Z Challenge — using o3/o4 mini and GPT 4.1 models to discover previously unknown archaeological sites: https://t.co/nVi4JghOIB
— Greg Brockman (@gdb) May 15, 2025
Today, Box announced new AI Agents to work with enterprise content, powering Deep Research, Search, and enhanced Data Extraction. There’s a tremendous amount of value that’s trapped in unstructured data, from contracts to research data, that we can finally unlock with AI. https://t.co/FYHxT4tDS5
— Aaron Levie (@levie) May 15, 2025
Here's the full workshop handout plus annotated slides from "Building software on top of Large Language Models", a three hour tutorial I presented yesterday at PyCon US #PyConUS https://t.co/P8tnEOEh0t
— Simon Willison (@simonw) May 15, 2025
Introducing Manus image generation. Manus doesn’t just generate images. It understands your intent, plans a solution, and knows how to effectively use image generation along with other tools to accomplish your task. https://t.co/pWbaVOy7Xb
— ManusAI (@ManusAI_HQ) May 15, 2025
Ben Kus, CTO of @Box is live talking about document extraction agents and how GenAI changed document processing! - Prototype worked incredibly well, then started facing challenges. Tasks and expectations became more and more complex, hit the "trough of disillusionment" - https://t.co/LQtrDKQwkU
— LangChain (@LangChainAI) May 15, 2025
This is bigger than you think DeepMind has unveiled an AI system called AlphaEvolve that uses Gemini models and evolutionary algorithms to independently find new ways to solve complex computational tasks - such as matrix multiplication, a cornerstone of any modern AI. Why is https://t.co/r3tnBDKjJr
— Chubby♨️ (@kimmonismus) May 14, 2025
Google DeepMind’s just dropped AlphaEvolve > a self-improving agent that combines the generative power of Gemini Pro and Flash with automated evaluators and an evolutionary algorithmic loop > generates complete algorithms in code, runs and evaluates them, and iteratively https://t.co/uvUVJanI9h https://t.co/jH3wsOeljp
— NIK (@ns123abc) May 14, 2025
People aren’t talking enough about how most of OpenAI’s tech stack runs on Python. https://t.co/0S0K77Wz3d
— Pietro Schirano (@skirano) May 14, 2025
Transformers visualized WITH CODE. https://t.co/mMqeeIIeQQ
— โฮตsam (@Hesamation) May 14, 2025
At first, I thought it was odd I was getting sell side notes from the chemical sector on DuPont. Then when you realize they have struck gold as technology chemicals company, this spin off makes a lot of sense. Also been in the works for a while. https://t.co/cq6tOo8l2k
— Ben Bajarin (@BenBajarin) May 14, 2025
Sequoia on AI's trillion-dollar opportunity: https://t.co/N7OoYA6msf
— Ben Lang (@benln) May 14, 2025
OpenAI introduces HealthBench, a new open-source LLM benchmark for health! Across frontier models, o3 is the best performing model with a score of 60%, followed by Grok 3 (54%) and Gemini 2.5 Pro (52%) A deeper dive: HealthBench consists of 5,000 synthetically generated https://t.co/rcQaEj3deI
— Tanishq Mathew Abraham, Ph.D. (@iScienceLuvr) May 12, 2025
Deep Research -> beautiful PDF: https://t.co/pTM2OCp7bJ
— Greg Brockman (@gdb) May 13, 2025
Approval AI (@approval_ai) is America’s first AI-powered mortgage platform, helping homebuyers get the best deal by comparing lenders, negotiating rates, and simplifying every step of the process. https://t.co/4zG6c4xfoV Congrats on the launch, @arjunlalwani and Helly! https://t.co/BC4iOo4U5l
— Y Combinator (@ycombinator) May 12, 2025
Tool-using LLMs can learn to reason—without reasoning traces. ๐ฅ We present Nemotron-Research-Tool-N1, a family of tool-using reasoning LLMs trained entirely via rule-based reinforcement learning—no reasoning supervision, no distillation. ๐ Paper: https://t.co/QGE4QVxXYX ๐ป https://t.co/cEl5GyTT1B
— Shaokun Zhang (@ShaokunZhang1) May 13, 2025
Claude 3.7 system prompt is huge! Weighing in at 110kb. No other LLM even comes close! Why is it this big? Is Claude very different from other LLMs? https://t.co/p63opf7TJL
— Carlos E. Perez (@IntuitMachine) May 13, 2025
This LLM from Meta skips the tokenizer, reading text like computers do (bytes!) for better performance. The training and inference code for BLT is released on GitHub. @AIatMeta just released the model weights for their 8B-param Dynamic Byte Latent Transformer (BLT), an https://t.co/iTDPbgyPG5 https://t.co/ohni3799lP
— Rohan Paul (@rohanpaul_ai) May 12, 2025
Bytedance, opensourced DeerFlow (Deep Exploration and Efficient Research Flow) LLM + Langchain + tools (web search, crawler, code exec). Its a community-driven Deep Research framework. https://t.co/xYjsvJZpzk
— Rohan Paul (@rohanpaul_ai) May 12, 2025
I really like @elevenlabsio's interactive landing page My favorite landing pages are the ones that drop you straight into the application and let you use it, you barely see that though I think it makes sense because why would you even need to sell to a customer with a big https://t.co/SL6aYVwhTu
— @levelsio (@levelsio) May 11, 2025
๐๐ธ๐ฝ https://t.co/LsU1TAkdQG
— LoA (@Lordofacca) May 10, 2025
Google literally dropped a 68-page Prompting Guide masterclass. https://t.co/soDRLHfmDh
— Hasan Toor ✪ (@hasantoxr) May 12, 2025
Super nice paper from Bytedance. With this, AI learns to code better by first learning how to pick its own high-quality training data. Releases Seed-Coder: Let the Code Model Curate Data for Itself Current LLMs for code pretraining heavily rely on manual data curation, which https://t.co/UH7IG5Vbyu
— Rohan Paul (@rohanpaul_ai) May 11, 2025
Btw to understand just how stable/fundamental these abstractions are, note that: since the open-sourcing of DSPy in Jan 2023, the fundamental way you code in it has only minimally changed syntactically. DSPy programs from July 2023 can still run today and be compiled with the https://t.co/SqIbegX15G
— Omar Khattab (@lateinteraction) May 11, 2025
Most influential LLM papers and the ideas they introduced (post 2017) A long thread ๐งต https://t.co/bFpbhjXsYK
— Pramod Goyal (@goyal__pramod) May 11, 2025
๐จBREAKING: The World’s first true agentic browser just launched. Introducing Fellou, an AI that handles everything online. Plans, researches, and executes — all in one tab. Here’s how it works: https://t.co/psoMZvWJN8
— Tulsi Soni (@shedntcare_) May 10, 2025
$PLTR ontology is the basis for infinite, autonomous labor and in the aggregate, companies that basically run themselves. This is why I think $PLTR will evolve to be a $20T company over time. https://t.co/9XNUtKsQRD
— Antonio Linares (@alc2022) May 10, 2025
The world's shortest YouTube Shorts guide you'll ever need... https://t.co/RKAY7L6FZ0
— igorm (@yt_igm) May 10, 2025
Every AI dev should know how to implement these types of agentic systems. There is so much value and new use cases to build from these patterns. The best part is when you combine these ideas. Check out my new 4.5 hr course (with code examples) to learn more: https://t.co/v2KoB3A4oJ
— elvis (@omarsar0) May 10, 2025
Founder lessons and wild stories from building a $250B software company with @vladdoes and a walk down memory lane featuring Steve Jobs, Larry Ellison, and Tony Robbins. One of my favorite convos. https://t.co/i9IQE1paTv
— Marc Benioff (@Benioff) May 10, 2025
This is the guide I wish I had - didn't hold back. Everything I know. Enjoy. https://t.co/SBupr1yL9L
— Hrishi (@hrishioa) May 9, 2025
This is literally the only book you need to master reinforcement learning from scratch https://t.co/4KZ5ywDW8t
— Aadit Sheth (@aaditsh) May 9, 2025
My short talk at @sequoia AI Ascent on how we at @OpenAI are attempting to flip the old "cherry-on-top" meme about "Scaling RL" https://t.co/nVmpGZ8yan
— Dan Roberts (@danintheory) May 9, 2025
Gemini 2.5 Pro is the workhorse model that's quietly powering a lot of what we build @every we finally wrote about it in today's Vibe Check https://t.co/ecZ8Q8RqK5
— Dan Shipper ๐ง (@danshipper) May 9, 2025
The way OpenAI talks about RL gets me so hard. Josh Tobin said they have the "recipe" to scale incredibly useful computer agents, and that "AGI" is pretty much just an operational problem now. https://t.co/P1itcbThap
— JB (@JasonBotterill3) May 9, 2025
A week in AI Agents is like a year in traditional software. Here is everything that happened this week in AI Agents from LlamaIndex, Cua, AgentStack, AgentOps, AgentMail, Agentforce, Mastra, Firecrawl, & more. ๐งต (save for later) https://t.co/8n5qxOIbYJ
— Adam Silverman (Hiring!) ๐️ (@AtomSilverman) May 9, 2025
OpenAI finally released the RL finetuning API today. I still stand by what I wrote about it 4 months ago while at some random NeurIPS cafe. https://t.co/UWmhn7Z0mE
— Nathan Lambert (@natolambert) May 8, 2025
Remember reinforcement fine-tuning? We’ve been working away at it since last December, and it’s available today with OpenAI o4-mini! RFT uses chain-of-thought reasoning and task-specific grading to improve model performance—especially useful for complex domains. Take https://t.co/7V8Oxlfa2L https://t.co/zmCzz7rE7J
— OpenAI Developers (@OpenAIDevs) May 8, 2025
Wild. With OpenAI’s new Image API, we’ve entered a new era where AI writes *and* designs your posts! ๐คฏ Nate Herkelman used it to turn @n8n_io into a full content + design studio: ↳ reads your post ↳ builds aligned visuals ↳ designs charts + graphics! Full demo in ๐งต ↓ https://t.co/3vprakCPxs
— Charly Wargnier (@DataChaz) May 8, 2025
My life got so much easier after watching this RL series, I can’t even explain. https://t.co/muPSqQgGxk
— pdawg (@prathamgrv) May 7, 2025
Anduril Industries CEO Palmer Luckey on Moonshots with Peter Diamandis: https://t.co/6LHpSFGudB
— vitrupo (@vitrupo) May 8, 2025
Box Founder Aaron Levie: Build for Customers Living in the Future "@paulg has a great article about living in the future and and building for what is missing when you're living in the future. If you find customers that are working in the future, you will be able to work with https://t.co/9StYPVjT72
— Overlap: Business & Tech (@Overlap_Tech) May 7, 2025
You do not want to miss the MCP launches & announcements, from Amazon, Google, Razorpay, Wiz, Baidu,CrowdStrike, Langchain, Shadcn, LlamaIndex, and more this week! I summarized everything you need to know below. (save for later) https://t.co/jb8MeS8OBg
— Adam Silverman (Hiring!) ๐️ (@AtomSilverman) May 7, 2025
MCP and Function Calling solve different problems. One tells WHAT tool to use, the other standardizes HOW tools connect. ๐ง๐ต๐ฒ๐'๐ฟ๐ฒ ๐ฐ๐ผ๐บ๐ฝ๐น๐ฒ๐บ๐ฒ๐ป๐๐ฎ๐ฟ๐ ๐๐ฒ๐ฐ๐ต๐ป๐ผ๐น๐ผ๐ด๐ถ๐ฒ๐, ๐ป๐ผ๐ ๐ฟ๐ถ๐๐ฎ๐น๐. ๐ ๐๐๐ป๐ฐ๐๐ถ๐ผ๐ป ๐๐ฎ๐น๐น๐ถ๐ป๐ด'๐ ๐ฐ๐ผ๐ฟ๐ฒ ๐ฝ๐๐ฟ๐ฝ๐ผ๐๐ฒ: 1. https://t.co/3sb3prGddU
— Victoria Slocum (@victorialslocum) May 7, 2025
The future of AI is AI systems being able to talk to each other to perform tasks. At Box, we’re all in on MCP to support more seamless AI experiences with documents, media, or any other unstructured data. https://t.co/lu9gld9qeJ
— Aaron Levie (@levie) May 7, 2025
๐Dynamic Graph Memory in @mem0ai - when facts needs relationships Mem0’s Dynamic Graph Memory turns every extracted fact into a node, wires the relationships automatically, and keeps the whole graph fresh as new updates come in. https://t.co/dBZnkdyDS5
— Taranjeet (@taranjeetio) May 6, 2025
https://t.co/Gn2hVjmgss
— Mario Thomas (@mariothomas) May 7, 2025
DeepSeek released another model: this time, it is able to do what most reasoning models can't. Below we explain how they did this ๐งต๐ 1/6 https://t.co/Vv40jhLQ6x
— SemiAnalysis (@SemiAnalysis_) May 6, 2025
To better understand MCPs and agentic workflows, I built news-agents to generate daily news recaps. The main agent spawns sub-agents, assigns them news feeds to parse and summarize, then generates a final overall summary plus analysis. Pretty cool at the 20-second mark! https://t.co/DinVaumUMp
— Eugene Yan (@eugeneyan) May 7, 2025
ChatGPT-4o can transform any google map view as ancient treasure map! Prompt below ⤵️ https://t.co/VxI7Az4ECP
— Umesh (@umesh_ai) May 6, 2025
Nvidia presents Llama-Nemotron: Efficient Reasoning Models - An open family of models w/ exceptional reasoning capabilities and inference efficiency - Discusses the training procedure, incl. NAS from Llama 3 for accelerated inference, knowledge distillation, and continued https://t.co/q40PAOeprw
— Aran Komatsuzaki (@arankomatsuzaki) May 5, 2025
NEW: MSFT has not yet agreed to OpenAI's restructure plans and is a key holdout in negotiations needed to get the restructure done (along w/ approval from state AG offices) per sources Still being decided how much equity MSFT gets from @dinabass + me https://t.co/iGHmQDzD4y
— Shirin Ghaffary (@shiringhaffary) May 6, 2025
https://t.co/ogBkLzU1qI
— Friedrich Nietzsche | Philosophy & Psychology ๐ง (@QuoteNietzsche) May 5, 2025
You must know these ๐๐ด๐ฒ๐ป๐๐ถ๐ฐ ๐ฆ๐๐๐๐ฒ๐บ ๐ช๐ผ๐ฟ๐ธ๐ณ๐น๐ผ๐ ๐ฃ๐ฎ๐๐๐ฒ๐ฟ๐ป๐ as an ๐๐ ๐๐ป๐ด๐ถ๐ป๐ฒ๐ฒ๐ฟ. If you are building Agentic Systems in an Enterprise setting you will soon discover that the simplest workflow patterns work the best and bring the most business value. https://t.co/tU3Kaje8kq
— Aurimas Griciลซnas (@Aurimas_Gr) May 5, 2025
Platform Power Is Underrated Platforms are so powerful that Apple’s latest court loss won’t change the course of the iPhone; that’s why it’s worth it. https://t.co/LKxqds75Im
— Stratechery (@stratechery) May 5, 2025
From Zero to $116K/mo in 6 months with Email Marketing - Playbook! ๐ฅ https://t.co/qwhwKunuCD
— Natia Kurdadze (@natiakourdadze) May 4, 2025
๐จ BIOMECHANICAL FUSION ๐จ Base prompt : [SUBJECT] portrayed in a Biomechanical Fusion, blending organic forms with advanced cybernetic technology. Employ high-contrast imagery with subtle accents of [COLOR1] and [COLOR2] to highlight the melding of flesh and machine Check https://t.co/nRCu6iiJtC
— LudovicCreator (@LudovicCreator) May 4, 2025
What if the only thing standing between you and your most confident self on stage was a pair of glasses? Introducing the Even G1 by @EvenRealities sleek, everyday eyewear that quietly houses something extraordinary: an AI-powered, invisible teleprompter that moves with your https://t.co/UyWlh9W22C
— Neel Shah (@codedailyML) May 4, 2025
This is next level. Nate Herkelman uses @n8n_io to bring full observability to AI agents! ๐คฏ His setup: ↳ logs all agents’ actions ↳ tracks models + token usage ↳ monitors costs ↳ records I/O + errors Full 9-minute walkthrough in ๐งต ↓ https://t.co/dwTxWqcZvj
— Charly Wargnier (@DataChaz) May 4, 2025
Redux of a Delaunay triangulation construction demo I made and posted on OpenProcessing in 2012. https://t.co/xXevAJM7XF
— Raven Kwok (@RavenKwok) May 4, 2025
I see abstract AI agent architectures everywhere. But no one explains how to build them in practice. Here's a practical guide to doing it with n8n: ๐งต https://t.co/b18DtBLq39
— Paweล Huryn (@PawelHuryn) May 4, 2025
How we built an AI agent that creates daily content from our existing material (n8n + Cloudflare + AutoRAG setup) We’ve been experimenting with autonomous content workflows inside our business—and one of the most useful agents we built is one that turns raw content into polished https://t.co/meQYYzMiJW
— Jay ⛽️ (@jelanifuel) May 3, 2025
