2024: Death of the Data Scientist 2025: Birth of the AI Data Scientist Start learning AI. It's virtually free. Just learn these: https://t.co/R5wpGXWV38
— 🔥 Matt Dancho (Business Science) 🔥 (@mdancho84) Feb 28, 2025
Generative vs. discriminative models in ML: Generative models: - learn the distribution so they can generate new samples. - possess discriminative properties—we can use them for classification. Discriminative models don't have generative properties. https://t.co/RiFHGjCngQ
— Avi Chawla (@_avichawla) Feb 28, 2025
Harper (@Harper359088579) is building the first AI-native commercial insurance brokerage. Since October, they've written over $6M in annualized premiums across 35 states. https://t.co/zy8raCtFN1 Congrats on the launch, @DakotahBlueRice and @nairtu965! https://t.co/dfsewnyhVe
— Y Combinator (@ycombinator) Feb 28, 2025
Google Veo-2 Minimax Hailuo Kuaishou Kling Alibaba Wan Luma Ray Runway Gen-3 Tencent HunyuanVideo Skywork SkyReels Pika Pika OpenAI Sora Stepfun Step-video Pixverse Genmo Mochi-1 Lightricks LTX-Video Haiper Zhipu CogVideo
— fofr (@fofrAI) Feb 28, 2025
McKinsey just revealed the extinction-level threat to business: Only 1% of companies have mastered AI technology. Early adopters are seeing 370% returns while others fall behind. Here's what's about to happen to the other 99%: https://t.co/Uz3O5X4L3M
— Kevin Henrikson (@KevinHenrikson) Feb 27, 2025
Agentic AI Search just had a DeepSeek Moment. This agentic search engine outperforms Perplexity Pro with only 3% of funding. It has Multi-agent architecture with advanced reranker built on custom Llama 3.3 models. 100% free to try. https://t.co/2i1yirFvIu
— Shubham Saboo (@Saboo_Shubham_) Feb 28, 2025
LLMs are automating data ETL end-to-end - and it starts with structured extraction. We’re excited to announce the launch of LlamaExtract 🧑🔬🤖: a GenAI-native extraction agent that adapts the latest models to offer accurate structured extraction over large amounts of complex… https://t.co/BOvZg0NKu5 https://t.co/2PVRw7OzQB https://t.co/BKgPU6mmQ0
— Jerry Liu (@jerryjliu0) Feb 27, 2025
New UI for AI Prompts already shipped. https://t.co/yLoNwR5Kuf https://t.co/scuOXEE3mW
— Kevin Kern (@kregenrek) Feb 27, 2025
You dot com's ARI (Advanced Research Intelligence) is shockingly good. I just had it run a query that I also tried on Grok 3 Deep Search and Open AI Deep Research. Let's check out how well it did! https://t.co/8kb2V2BRIT
— Carlos E. Perez (@IntuitMachine) Feb 27, 2025
this ai agent replaces a $100K/year growth manager and creates 100 videos a day I built a fully-autonomous ai agent that repurposes winning ads on autopilot. I talked with dozen of marketers from.. - top YC startups - hyper-scaling DTC brands - $8fig ads agencies They all… https://t.co/U1R4slhi4b https://t.co/XuhAgqw03r
— Romain Torres (@rom1trs) Feb 27, 2025
A little animation from an upcoming project https://t.co/tvcCBQLi6C
— Clément Moreau (@clementmoreau3d) Feb 25, 2025
This was actually quite tricky to implement. The servers are replicated. We wanted to use REST instead of websockets. We couldn't use sticky routing as that would hurt graceful restarts. You need to carefully avoid race conditions between the client state and the token stream. https://t.co/ctrOoQcEh3
— Toby Pohlen (@TobyPhln) Feb 26, 2025
this prompt is amazing for Cursor "Reflect on 5-7 different possible sources of the problem, distill those down to 1-2 most likely sources, and then add logs to validate your assumptions before we move onto implementing the actual code fix" thanks Ted! https://t.co/VqMy8kSTJ4
— Filip Kowalski (@filippkowalski) Feb 26, 2025
John Carmack is just out there quietly building AGI. https://t.co/lHUNbz4xUx
— Packy McCormick (@packyM) Feb 26, 2025
.@BenBajarin and I wrote this before the iPhone 16e but I think this is a better time than ever to re-read it to understand why an iPhone like the 16e and Apple Intelligence is going to do really well in Asian markets https://t.co/vCZIIx5yzf
— Max Weinbach (@MaxWinebach) Feb 27, 2025
$AMZN CEO JUST SAID AMAZON HAS BUILT OR IS BUILDING 1,000+ GEN AI APPS 😳 https://t.co/Cx8IWi63re
— Shay Boloor (@StockSavvyShay) Feb 26, 2025
New this week: All-new Figma plugin 💫 Meet Jitter’s revamped Figma plugin — with new import options powered by the infinite canvas! 📦 Import @figma Slides decks into Jitter 🖱️ Multi-select layers & frames 🪄 Import your entire Figma file at once Learn more and try it now 👇 https://t.co/Dj0BEj5DEi
— Jitter (@jittervideo) Feb 26, 2025
MoBA, Mixture of Block Attention, from @Kimi_Moonshot improves handling long-context tasks with no fixed attention patterns. Applying ideas from Mixture of Experts (MoE) to attention, MoBA lets the model dynamically decide where to focus. This allows MoBA to be 6.5x faster than… https://t.co/gkeXEz5uZa https://t.co/TnkgatoXaP
— TuringPost (@TheTuringPost) Feb 26, 2025
🚨 Gemini Code Assist is now free for individual use. No cloud account needed, virtually no usage limits. 128k token input. This excellent IDE extension for Visual Studio Code and Jetbrains IDEs use Gemini 2.0 and compliments @googleaistudio in your AI toolbox. Try it! https://t.co/38Kuv1ObS2
— Richard Seroter (@rseroter) Feb 25, 2025
the @cursor_ai folks got so much right, so early. 1.5 years ago, before anyone had really heard of cursor, we were building a really slick vscode extension called "sidekick" -- think superhuman but for coding. but while we decided to make a vscode extension, they decided to… https://t.co/GUxO6p8Dyx https://t.co/3fXdZBXvcx
— ben (@benhylak) Feb 25, 2025
2025 could be the year of Deep(Re)Search. Test-time compute and reasoning model are transforming search systems now. With <think>, users have been educated to accept delayed gratification—longer waiting times in exchange for higher-quality, actionable results, much like the… https://t.co/v3qC703u5u https://t.co/UnR788M9oh
— Jina AI (@JinaAI_) Feb 25, 2025
We’re making Deep Research available as an endpoint to all developers through the Perplexity Sonar API to help people build their custom research agents and workflows! Excited to see what people are going to build using this! https://t.co/6hU7fIRFqV
— Aravind Srinivas (@AravSrinivas) Feb 25, 2025
Less than 5% of books are released as audiobooks. Today we’re changing that with ElevenReader Publishing by @elevenlabsio. It’s the fastest, zero-cost way for authors to convert their books into immersive audiobooks with built-in distribution globally on the ElevenReader iOS… https://t.co/maNBpuCDG3 https://t.co/z5XdZ9CR0M
— ElevenReader (@elevenreader) Feb 25, 2025
Transformers vs Mixture of experts in LLMs visually explained: https://t.co/eDwKJhoZvq https://t.co/ryezDRuaOb
— Daily Dose of Data Science (@DailyDoseOfDS_) Feb 25, 2025
A First-Principles Approach to Agentic Intelligence https://t.co/cEOa8rJW0i
— Carlos E. Perez (@IntuitMachine) Feb 25, 2025
Claude 3.7 Sonnet is the best model in the world for code. It’s like having a world-class dev with exceptional taste. “build a next.js SaaS marketing template” And boom - 26 files of beautiful code in one shot. Absolutely unreal. https://t.co/vmH0tiVY7v
— Mckay Wrigley (@mckaywrigley) Feb 24, 2025
Self improving AI means *as people go through* the flows, AI is running its own evals and thinking of ways to improve performance, adding to the script and making changes Leaping is the only one doing it right now, and it's a sight to see with 10,000 customer calls per day https://t.co/sV6YPqRQw0
— Garry Tan (@garrytan) Feb 24, 2025
https://t.co/O81rtjF0Al
— David Deutsch (@DavidDeutschOxf) Feb 24, 2025
Here's the write-up https://t.co/sgqnChgy6h
— Ethan Mollick (@emollick) Feb 24, 2025
Biological Cognition and the Future of AI: Agential Matter, Nano-Intentionality, and the Limits of… https://t.co/qaH9lnz6ik
— Carlos E. Perez (@IntuitMachine) Feb 24, 2025
Google's Co-Scientist in Simple Terms The Generation Agent is like the lab's theoreticians - smart thinkers who come up with initial novel hypotheses by synthesizing insights from the scientific literature and debating ideas among themselves. The Reflection Agent acts as the… https://t.co/g5lepkYdaW https://t.co/UrNLWFMGdz
— Carlos E. Perez (@IntuitMachine) Feb 24, 2025
https://t.co/pmHKaxMDE7
— Gopi Vikranth (@GopiVikranth) Feb 23, 2025
NEW N8N AI Agent Update is INSANE 🤯 Want the SOP? DM me. https://t.co/WjtlXxrwrd
— Julian Goldie SEO (@JulianGoldieSEO) Feb 23, 2025
TD COWEN CHANNEL CHECKS INDICATE MICROSOFT CANCELING U.S. DATA CENTER LEASES, AMID POTENTIAL OVERSUPPLY Analyst comments: "Our channel checks indicate that $MSFT has 1) canceled leases in the US, totaling 'a couple of hundred MWs' with at least two private data center operators,… https://t.co/EKP04yMVAi https://t.co/eXhavvrZl9
— Wall St Engine (@wallstengine) Feb 23, 2025
time for a paper and code review! this work claims to be both a new framework and a new benchmark for AI research agents! for such notable claims, it's worth digging into a bit, starting with Table 1: the columns here are interesting. it seems like "Agentic Harness" was added… https://t.co/vuJ9Ep9jir https://t.co/OY1KbZtSkI https://t.co/zQSdwsUMM0
— Susan Zhang (@suchenzang) Feb 23, 2025
Here are the top AI Papers of the Week (Feb 10-16): - AI Co-Scientist - Open-Reasoner-Zero - The AI CUDA Engineer - Native Sparse Attention - The Danger of Overthinking - Large Language Diffusion Model Read on for more:
— DAIR.AI (@dair_ai) Feb 23, 2025
wow... this new AI is incredible. Kaiber AI just dropped film Superstudio Pro, a node based canvas that can generate storyboard, video, voice, sfx(MMAudio) and even... train character, object, place and style all in one page step by step tutorial: https://t.co/H0TtFvVuSv
— el.cine (@EHuanglu) Feb 23, 2025
Warren Buffett who is currently the 9th richest person in the world just sent out this full letter to Berkshire Hathaway shareholders sharing his most recent thoughts A thread 🧵 https://t.co/Ch9r6auZx4
— Evan (@StockMKTNewz) Feb 22, 2025
China's AI is getting crazier.. ByteDance’s new AI model, OmniHuman-1 can generate video with realistic lip sync even if its side view and... it can control expression and gesture using audio and text.. 10 examples: https://t.co/EGZ8FXxduN
— el.cine (@EHuanglu) Feb 22, 2025
There's lots of people talking about GPT wrappers, not enough talking about voice AI wrappers. So, I researched 25+ voice AI startups and wrote up what I learned in 1 page. https://t.co/1MbQqkXpV4
— GREG ISENBERG (@gregisenberg) Feb 22, 2025
Chain-of-Agents (CoA) allows AI worker agents to collaborate on unlimited amounts of text, with a manager agent summarizing the results. Here are CoA's benefits and why it's better than RAG: ▪️ Performance improves as input length increases (with 400k tokens, it achieves up to… https://t.co/cYeG3n7E22 https://t.co/1pwL6OXawW
— TuringPost (@TheTuringPost) Feb 22, 2025
This app is crazy. It claims to work like a drug—but with sound. "Moongate: Binaural Beats" is making $600K/month with 200K downloads, using science-backed frequencies to help users sleep, focus, and relax. Here’s how they built a subscription machine. (1/11)🧵 https://t.co/REqSVz9CDb https://t.co/D9oMiuLusu
— Siro (@Siron93) Feb 21, 2025
Introducing the judges for the ElevenLabs x a16z WorldWide hackathon this weekend: Mati Staniszewski - Co-Founder, @elevenlabsio Guillermo Rauch - CEO, @vercel James Hawkins - CEO, @posthog Bryan Kim - Partner, @a16z Justine Moore - Partner, @a16z Anton Osika - Co-Founder,… https://t.co/nOheqLmO6z https://t.co/oLWiIZ3Vbp
— Louis Jordan (@louisjoejordan) Feb 21, 2025
WOW — the new SkyReels video model allows for really precise editing via FlowEdit. The top is the original video, the middle is my last attempt that required training an entire LoRA (extra model), and the bottom generation with the new model and a single image! https://t.co/ymyMlNbZsj
— Zack Abrams (@ZackDAbrams) Feb 19, 2025
it’s been faster to sell AI apps in this tech cycle as opposed to cloud apps in the last one https://t.co/95lLL16pJX
— Diana (@sdianahu) Feb 20, 2025
The Download: Microsoft’s quantum chip, and explaining rising energy demand https://t.co/x5skjwVDiR
— MIT Technology Review (@techreview) Feb 20, 2025
Today Gumloop rolls out 𝘭𝘪𝘵𝘦𝘳𝘢𝘭 magic ✨ Our new AI Web Research node finds you the answers to any question by scouring the web. -Is this company SOC2 compliant? -What university did this person go to? -What talks did they give? Millions of new use cases unlocked 🔓 https://t.co/NcE0JewuxU
— Gumloop (@gumloop_ai) Feb 20, 2025
One-click AI agents that cost $0 to build and deploy? This is insane. 🤯 Key points from the video: • Build custom AI chatbots using DeepSeek's open-source models • Add knowledge bases from URLs, Wikipedia, or YouTube • Create content automation pipelines with memory… https://t.co/9ptUOQ3vh1 https://t.co/niKnB13pTV
— Julian Goldie SEO (@JulianGoldieSEO) Feb 20, 2025
Open Deep Research Deep research is one of the most popular agent use-cases. Here is an open deep researcher w/ ability to configure report structure, planner/writer LLMs, search APIs, search depth, etc. 📽️ https://t.co/JbrWTiBewN Code: https://t.co/ABYRrnawXC https://t.co/vNtiZWHPEx https://t.co/4R9GoQz5K8
— LangChain (@LangChainAI) Feb 20, 2025
God Mode UX: Why Your Next Interface Will Look More Like StarCraft Than Slack 👁️ It’s time to zoom out and see our AI from above 🛰️ because herding a swarm of digital agents needs more than a single chat window. ✨ Read more here ⬇️ https://t.co/PWVH9Q3c1d https://t.co/zqubtHU5F1
— 𝙳𝚊𝚗𝚒𝚎𝚕 ☈ (@sadasant) Feb 18, 2025
Tableau is about to die. Introducing PandasAI, a free alternative for fast Business Intelligence. Let dive in: https://t.co/M6yzkWwARQ
— 🔥 Matt Dancho (Business Science) 🔥 (@mdancho84) Feb 19, 2025
I tested Grok 3 and DeepSeek v3 with same critical prompts. The results will blow your mind. Grok 3 Vs. DeepSeek v3 (Video demos are included) https://t.co/t5Vpp8SvmD
— Alex Prompter (@alex_prompter) Feb 18, 2025
A directed graph of every Chess opening https://t.co/tIJuOkAVVU
— Anthony Bonato (@Anthony_Bonato) Feb 19, 2025
A new AI co-scientist, powered by Gemini 2.0, a glimpse into the future… https://t.co/YsEO5FH3uy
— Logan Kilpatrick (@OfficialLoganK) Feb 19, 2025
Game over! https://t.co/TQoeKamSea
— Carlos E. Perez (@IntuitMachine) Feb 19, 2025
🚀 Game-Changer for AI Agents! 🇨🇳🔥 China's latest RAG breakthrough is here! After DeepSeek R1, a new open-source research unlocks near-infinite context for LLMs—no more limits on processing entire codebases & docs! 📜➡️🤖 How? 👇 🔹 Mixture of Experts + Sparse Attention →… https://t.co/Ta3viqsraP https://t.co/jbjq4l3WEK
— Mervin Praison (@MervinPraison) Feb 19, 2025
AI Agents 101: 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗠𝗲𝗺𝗼𝗿𝘆. In general, the memory for an agent is something that we provide via context in the prompt passed to LLM that helps the agent to better plan and react given past interactions or data not immediately available. It is useful to group… https://t.co/rnQf4dUPnT https://t.co/dzPPW3wWPm
— Aurimas Griciūnas (@Aurimas_Gr) Feb 19, 2025
@satyanadella Congrats!! One correction: quantum adiabatic computing and supremacy for useful apps is not decades away. It is already here with D-Wave $QBTS https://t.co/SQ2BOlF9da
— Steve Jurvetson (@FutureJurvetson) Feb 19, 2025
Read more about our discovery, and why it matters, here: https://t.co/fzeVX6G95w
— Satya Nadella (@satyanadella) Feb 19, 2025
A couple reflections on the quantum computing breakthrough we just announced... Most of us grew up learning there are three main types of matter that matter: solid, liquid, and gas. Today, that changed. After a nearly 20 year pursuit, we’ve created an entirely new state of… https://t.co/WCLn20Ugmi https://t.co/Vp4sxMHNjc
— Satya Nadella (@satyanadella) Feb 19, 2025
New Substack article up - explaining why DataRepublican dotcom was necessary to expose the awards spending, and why critics who say the data was already available are ignorant: Data Isn’t Transparency: Why USASpending dotgov Buried the Truth (and DOGE Dug It Up) Link… https://t.co/1k6VSDi7Ez https://t.co/JSOeJDsesQ
— DataRepublican (small r) (@DataRepublican) Feb 19, 2025
Open-source AIOps and alert management platform https://t.co/UiSD7snkxO
— Tom Dörr (@tom_doerr) Feb 18, 2025
A gift to the world. Amazing. 🇺🇸 https://t.co/tFcqAcLHaq
— Marc Andreessen 🇺🇸 (@pmarca) Feb 18, 2025
4 months ago, this was just a dream. today, it’s real. meet Magic – an MCP AI agent that understands when you need a component and crafts it for you, inspired by real components published on 21st[.]dev & @shadcn /ui. 🚀 free during beta week 💰 50% revenue goes to 21st[.]dev… https://t.co/wW0WA8GZAh https://t.co/YamCxK0Dyy
— serafim | 21st.dev | e/shadcn (@serafimcloud) Feb 17, 2025
I've been using Windsurf 3 for the past few days and really loving it. This is a great tutorial on how to use MCP in Windsurf 3, definitely worth watching. It's a really powerful. https://t.co/EQQR11N2dD
— AshutoshShrivastava (@ai_for_success) Feb 18, 2025
Announcing DeepSeek R1 Documentation Assistant 🤖 Scrape any documentation with @firecrawl_dev, then chat with your docs using DeepSeek R1 and @ollama. Zero cloud LLM dependency, 100% local! Built with @Streamlit for a seamless experience ⚡ https://t.co/LLY7w13Vv1
— Caleb Peffer (Hiring!) (@CalebPeffer) Feb 17, 2025
Agentic Commerce is going to be a big deal this year. Coinbase Commerce (@CommerceCB) can help get your shopping cart AI Agent enabled. And @CoinbaseDev can help give your AI Agent a wallet. Get in touch. Let's make this happen. https://t.co/em5w1chy6a
— Brian Armstrong (@brian_armstrong) Feb 17, 2025
Transform PDFs to audio podcasts https://t.co/Na4nt6RNLV
— Tom Dörr (@tom_doerr) Feb 17, 2025
The Deep Research paradigm highlights what AI Agents will look like for knowledge work. You’ll kick off unlimited Agents that are all doing tasks on your behalf and then you review and collate the final result. Projects that took days or weeks will take a couple of hours.
— Aaron Levie (@levie) Feb 17, 2025
https://t.co/G1kEhRGZIl It's nuts what Midjourney and Luma Ray 2 can achieve today already. Mark my words: Animes will be the first one that are 100% AI generated
— Chubby♨️ (@kimmonismus) Feb 17, 2025
Microsoft silently updated OmniParser on the hub 👀 60% faster than v1 - sub-second latency on a 4090! "OmniParser is a general screen parsing tool, which interprets/converts UI screenshot to structured format, to improve existing LLM based UI agent." Bonus: you can try it… https://t.co/N9nm8C9aaB https://t.co/VcKcewb7nG
— Vaibhav (VB) Srivastav (@reach_vb) Feb 17, 2025
NEW 1-Click DeepSeek AI Agents are INSANE! 🤯 https://t.co/FzEEhYJhJ7
— Julian Goldie SEO (@JulianGoldieSEO) Feb 17, 2025
The 12 Steps Defining the New Digital Economy 1. $ASML designs and manufactures advanced photolithography machines essential for semiconductor chip production -- used by $TSM to fabricate high-performance semiconductor chips. 2. $ARM designs the fundamental architectures that… https://t.co/iCkP8A9iQG
— Shay Boloor (@StockSavvyShay) Feb 16, 2025
Microsoft just dropped OmniParser V2, this changes everything. This AI sees your screen, understands it, and takes action, just like a human. 100% free & open source! https://t.co/lOwdvICClE
— Min Choi (@minchoi) Feb 15, 2025
I really like MCP by @AnthropicAI. It aims to be the USB-C of llm tools. However, it's incomplete right now: To follow the analogy, right now its cable and wire protocol. It's missing the plug. I think MCP should suggest a textual standard of pointing the LLM to a MCP endpoint… https://t.co/rR2afFrRcY
— tobi lutke (@tobi) Feb 16, 2025
"Mediocre youtube engineer" casually drops video on "killer" defence tech maybe without knowing it. Doing this with SDR and you can "see" things emitting a radio signal (all electronics) on a battlefield. Freeing information and not compartmentalized in corps/countries changes… https://t.co/as6pHJcNnC https://t.co/pY2MWxUwW2
— mariusheier (@mariusheier) Feb 16, 2025
OpenAI just dropped a paper that reveals the blueprint for creating the best AI coder in the world. But here’s the kicker: this strategy isn’t just for coding—it’s the clearest path to AGI and beyond. Let’s break it down 🧵👇 https://t.co/NJwbq4kxRs
— MatthewBerman (@MatthewBerman) Feb 16, 2025
Wow. Got a few hundred sci-fi movie recs here... The most recommended were: ▸ Forbidden Planet (10) ▸ Contact (9) ▸ Fifth Element (9) ▸ Event Horizon (8) ▸ Starship Troopers (8) ▸ Terminator 1 (8) ▸ The Thing (8) ▸ Terminator 2 (7) ▸ The Abyss (7) ▸ Planet of the Apes… https://t.co/8rF4dfjcHD https://t.co/41sb0gTXyJ
— Christian Keil (@pronounced_kyle) Feb 16, 2025
Lake Como, Italy 🇮🇹 https://t.co/OXJj115vgA
— Appart Design (@AppartDesign) Feb 14, 2025
🤖 🖥️ Clevrr Computer An intelligent automation agent that combines multi-modal AI with system control. Uses LangChain's PythonREPLAst and agent framework to enable AI-powered screen understanding and precise computer interactions. Check it out 🚀 https://t.co/hvgaUr2Q81 https://t.co/KuJQ89cCAT
— LangChain (@LangChainAI) Feb 15, 2025
AI-controlled cloud computer https://t.co/LBKsMljxfZ
— Tom Dörr (@tom_doerr) Feb 15, 2025
Narai-juku is a historical post town located in the Kiso Valley of Nagano Prefecture, Japan. This is a beautifully preserved Edo Period (1603-1868) neighborhood in Narai-juku. [📸 ag. lr.88] https://t.co/xoiZfif4Yl
— Massimo (@Rainmaker1973) Feb 15, 2025
I've put all my best travel tips from 53 years of traveling into one document. Enjoy! https://t.co/N46gsoD3hR
— Kevin Kelly (@kevin2kelly) Feb 14, 2025
@MdBreathe Here is a list of the Tennessee compounding pharmacies. TN does not require a doctor prescription for IVM. https://t.co/pM8TXSMbS2
— FlyingPatriot17 (@FlyingPatriot17) Feb 14, 2025
The hardest thing to do is start. https://t.co/JRaGX4NIXI
— Skai. (@superSaiyanSkai) Feb 14, 2025
studying the Medici family in-depth was one of the biggest historical wake-up calls for me i used to believe that history followed cycles/trends more than individuals. i no longer think so, indeed i believe the opposite to be true a handful of people can change history forever https://t.co/WM4dMVuqWW
— spor (@sporadicalia) Feb 13, 2025
Someone please open-source this. Everyone in the world deserves high quality education. https://t.co/XS9VgAkEIg
— Aadit Sheth (@aaditsh) Feb 13, 2025
These kids cooked a full planning system on Groq to have Richard Feynman teach you in your style. They have 17k users. Investors need to talk to these guys asap. https://t.co/5K9y6cea9L
— Ray Fernando (@RayFernando1337) Feb 14, 2025
Just tried @Hailuo_AI's new effects feature and it's so good, look out for a guest appearance at the end 😆 https://t.co/R0HbGwFN1i
— madpencil_ (@madpencil_) Feb 13, 2025
One of the first use cases people discovered with LLMs was marketing copy generation (e.g. Jasper/Copy), but the way it started was very much one-shot content generation. But the world is going towards marketing agents as sophisticated multi-step, hierarchical systems that also… https://t.co/CiXNLOCoIw https://t.co/fsmoRcNjNo https://t.co/vO6jfLvD8a
— Jerry Liu (@jerryjliu0) Feb 14, 2025
2-13-2025 An Interview with Uber CEO Dara Khosrowshahi About Aggregation and Autonomy https://t.co/AkXcJyswqb
— Stratechery (@stratechery) Feb 13, 2025
Github 👨🔧: An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models. The goal of this repo is to provide the simplest implementation of a deep research agent - e.g. an agent that can… https://t.co/tURDf8w2TO https://t.co/YPsJdgLoDw
— Rohan Paul (@rohanpaul_ai) Feb 13, 2025
📢 New @a16z thesis: AI will unbundle the BPO and disrupt the $300b outsourcing market. Enterprises often outsource important, but high-volume and repetitive work to BPOs. These BPOs rely heavily on human labor, often leading to slow turnaround times, human error, and… https://t.co/xYfwfbIW59 https://t.co/o7jcg8lyWx
— Kimberly Tan (@kimberlywtan) Feb 13, 2025
Less than a year after their Series A, @FAL has raised a $49M Series B, led by @notablecap and @BessemerVP, alongside @a16z, @KindredVentures, and @firstround Congrats @burkaygur, @gorkemyurt, @isidentical! Here’s a short (Turkish!) video of Gorkem explaining how they scaled to… https://t.co/maMQO6ZhpL https://t.co/0pgFSuR9GA
— Enis Hulli🎙 (@EnisHulli) Feb 12, 2025
Rocketship is an overused term. But it’s hard to come up with another way to describe @FAL’s trajectory — going from $1M to $40M ARR in one year is more vertical line than hockey stick. Their crazy focus on performance and reliability have meant that they’ve built what… https://t.co/Efr3tBASqs https://t.co/KKxpQGWCWP
— Todd Jackson (@tjack) Feb 12, 2025
Today we're sharing a major update to the Model Spec—a document which defines how we want our models to behave. The update reinforces our commitments to customizability, transparency, and intellectual freedom to explore, debate, and create with AI. https://t.co/EPbqDp0Sdj
— OpenAI (@OpenAI) Feb 12, 2025
Leonardo da Vinci on excellence: https://t.co/s8EquYSD16
— Reads with Ravi (@readswithravi) Feb 12, 2025
but most importantly, just start ‼️ https://t.co/oD4plbl87t
— Reads with Ravi (@readswithravi) Feb 12, 2025
1. Open-Deep-Research by HuggingFace https://t.co/WG5oNLuXc6
— Min Choi (@minchoi) Feb 11, 2025
Baubap (@Baubapmx) — AI underwriting ~$120M RR, ~3x YoY, profitable Been around a bit longer, but only raised $3M! https://t.co/OjEuqbnsmd
— Arfur Rock (@ArfurRock) Feb 11, 2025
OpenAI is getting Deepseek’d again. @Convergence_ai_, a tiny London startup just built one of the most capable AI agents for the web. Proxy is outperforming Operator on every benchmark handpicked by @OpenAI. → https://t.co/s8Xhf1ZQpu Let’s dive in! 🧵↓ https://t.co/wknRw9XW0P
— Charly Wargnier (@DataChaz) Feb 11, 2025
Carl is a Bitcoin core developer with years of experiencing building VPNs that can punch through any firewall — including Great ones. I don’t announce too many investments, but I think this helps advance Internet freedom worldwide. https://t.co/6kyRsfI8uV
— Balaji (@balajis) Feb 11, 2025
1/n Let me tell you why there's some air-tight logic behind Yann LeCun's conjecture. 🧵 https://t.co/Tg39KfUsk7
— Carlos E. Perez (@IntuitMachine) Feb 11, 2025
Introducing Gemini Cursor ✨ – a second multimodal AI cursor for your desktop that's open-source and free! Link below 👇 This experiment 🧪 reimagines how we interact with our computers because visual cues 👀 help us make sense of what we see on a screen. In this demo, I had my… https://t.co/hrZJPOFOAf https://t.co/q8xFlPX0Le https://t.co/lFhX8F89Xl
— Sriraam (@27upon2) Feb 11, 2025
openai charges $200/month for deep research i made it free and open source 👇 https://t.co/syg0nBXSzT
— Fekri (@fekdaoui) Feb 10, 2025
AI is great for quickly generating UI, but it’s hard to easily tweak and get the details right once it’s been generated. @onlookdev solves this with a visual editor that makes creating and editing React apps way faster. Congrats on the launch @FlyaKiet and @D_R_Farrell! https://t.co/f7aPVw3Gdx
— Aaron Epstein (@aaron_epstein) Feb 10, 2025
What the heck, this is an AI-generated voice and open source too under Apache 2.0 license. Damn! 🔥🔥 Zonos Beta is a new open-source highly expressive TTS model with high-fidelity voice cloning from @ZyphraAI Try it here 👇 https://t.co/GGn877CzGp
— AshutoshShrivastava (@ai_for_success) Feb 10, 2025
Andrew Ng and his team introduced Agentic Object Detection! No training needed, just describe an object, and AI thinks, plans, and finds it like OpenAI o1/o3. 6 examples (and how to try): 1. Detected occupied table https://t.co/KzPeQsvg6V
— Min Choi (@minchoi) Feb 10, 2025
Wow that's very impressive Zonos is a 100% open source AI model that can clone any voice 🤯 You can basically run it anywhere as it's only 1.6B parameters. Link below https://t.co/yztyL46NvN
— Paul Couvert (@itsPaulAi) Feb 10, 2025
Deep Research and Knowledge Value Deep Research is an AGI product for certain narrow domains; it’s ability to find anything on the Internet will make secret knowledge all the more valuable. https://t.co/1XyO76ntNw
— Stratechery (@stratechery) Feb 10, 2025
Mount Fuji, Japan. https://t.co/HuY1GDIJpP
— JAPAN 🇯🇵 (@Japantravelco) Feb 9, 2025
In depth explanation + examples of why sometimes you don't need rag (especially with gemini 2.0) https://t.co/RmRiGjwGkE
— Sully (@SullyOmarr) Feb 9, 2025
🔍 Deep Research Agent Build an autonomous research agent that processes Wikipedia content using LangChain's RAG with Milvus vector storage. Features intelligent question analysis, content synthesis, and local deployment. Ready to build your own? Check it out 🚀… https://t.co/lTk9OC4EZd https://t.co/tJZVrtSrjZ
— LangChain (@LangChainAI) Feb 9, 2025
Three Observations: https://t.co/Ctvga5vfMy
— Sam Altman (@sama) Feb 9, 2025
Netflix founder Reed Hastings on why a professional sports team is a good metaphor for a company: "Athletes on professional teams: - Demand excellence, counting on the manager to make sure every position is filled by the best person at any given time. - Train to win, expecting… https://t.co/WkXpgSsALr https://t.co/pQ2aH1LD0b
— Startup Archive (@StartupArchive_) Feb 9, 2025
Ingesting Millions of PDFs - Gemini Flash 2.0 finally solves it. Huge implecations for corporate usecases. https://t.co/1jyx3muc4L https://t.co/jS9rWNxqFm
— Rohan Paul (@rohanpaul_ai) Feb 8, 2025
NotebookLM will let users download mindmaps as images too 👀 * This feature is not available yet https://t.co/Igtybcei3F https://t.co/4Rn3qegnXk
— TestingCatalog News 🗞 (@testingcatalog) Feb 7, 2025
RAG is dead; long live RAG. https://t.co/4QbTvDXeGr https://t.co/EvGja3gBd0
— Jo Kristian Bergum (@jobergum) Feb 7, 2025
RT @Instalnfluencer: @clay_gtm @HeyGen_Official Try https://t.co/7raXwsxudL - affordable custom influencer videos in just a couple of steps…
— Gopi Vikranth (@GopiVikranth) Feb 8, 2025
@clay_gtm @HeyGen_Official Try https://t.co/7raXwsxudL - affordable custom influencer videos in just a couple of steps https://t.co/hB3jGas9zw
— InstaInfluencer (@Instalnfluencer) Feb 8, 2025
chains of thought for o3-mini! (we try to organize the raw CoT to make it more readable, and optionally to translate languages, but we try to keep it quite faithful to the raw one) great work from @mia_glaese, @joannejang, @akshaynathan_ , and their teams! https://t.co/A8JZIj45Lt
— Sam Altman (@sama) Feb 7, 2025
The future of document processing and workflows will rely heavily VLMs and LLMs. The best way to benchmark that is to throw the hardest documents you can find at the latest models. In this example, I try out feeding an electronics technical datasheet - with complex schematics… https://t.co/W18IgddN7L https://t.co/jmvYvfhCcl https://t.co/nHdnAKq03t
— Jerry Liu (@jerryjliu0) Feb 8, 2025
Is this what goes on in the "mind" of o3-mini? That's a lot of thoughts for a simple response! https://t.co/MWdNYYKadk
— Carlos E. Perez (@IntuitMachine) Feb 6, 2025
With Gemini 2.0 Flash you can really test the limits of what LLMs can do today. This is a great post for anyone working with large amounts of PDFs. https://t.co/GFXG9HVWxU
— elvis (@omarsar0) Feb 6, 2025
We used Gemini 2 Flash to build Cursor for arXiv papers Highlight any section of a paper to ask questions and “@” other papers to quickly add to context and compare results, benchmarks, etc. https://t.co/2KKTuDuzRh
— alphaXiv (@askalphaxiv) Feb 6, 2025
Browser Use UI can do DeepResearch👀 Repo is 100% open source ↓ Thanks @Gradio for powering the UI🔥 https://t.co/gue2CCo9Ut
— Gregor Zunic (@gregpr07) Feb 6, 2025
Deep Research from OpenAI seems like the most interesting AI product advance in a while. I'm really liking the outputs I'm getting so far. I wonder how much of the advance is about real-time access to the complete text of web pages, and how much is about more extensive CoT.
— John Collison (@collision) Feb 6, 2025
I'm excited to introduce my AI Exploratory Data Analysis (EDA) Agent that built EDA reports, performs correlation analysis, and missing data analysis in 20 seconds. Today, I'll share with you how to automate creating EDA reports with the AI EDA Agent, which is available on… https://t.co/P2fkq0Py4W https://t.co/CT7tNELUhh
— 🔥 Matt Dancho (Business Science) 🔥 (@mdancho84) Feb 6, 2025
Honored that @karpathy used Together AI to use DeepSeek R1 in his latest video! https://t.co/W5IKO034OK
— Together AI (@togethercompute) Feb 6, 2025
We have created a list of Cursor rules you can copy and paste into your codebase! Each rule comes with a sample description and glob pattern so that your rule is only applied to the appropriate use case. What other sample rules should we create on top of what we have now? https://t.co/M0wc2Y6cz7
— Tyler Shukert (@dshukertjr) Feb 6, 2025
I've said this so many times since last year, you can't ignore the multimodal capabilities of Google Gemini. It performs exceptionally well with PDFs and videos. No one, literally no one, comes close to Gemini in those two areas. https://t.co/XbI1WRlNAL https://t.co/WHz9FUuReo
— AshutoshShrivastava (@ai_for_success) Feb 6, 2025
$PLTR integrates Grok into AIP -- bringing xAI to the enterprise 👀 https://t.co/xOywlSctrE
— Shay Boloor (@StockSavvyShay) Feb 6, 2025
How to Milk OpenAI's Deep Research for All It’s Worth (Without Losing Your Marbles) The Juicy Bits: Why OpenAI Deep Research Is Awesome 1. Data-Glutton Extraordinaire Deep Research is like that friend who never stops eating at an all-you-can-eat buffet. It’s been fed more data… https://t.co/t7Dfu6GTUP https://t.co/cIN2WfAdgR
— Carlos E. Perez (@IntuitMachine) Feb 6, 2025
Chain-of-Thought (CoT) prompting struggles with complex spatial tasks. Existing multimodal CoT methods rely on external tools or simplified text, limiting expressiveness. This paper introduces Multimodal Visualization-of-Thought (MVoT), enabling LLMs to generate image… https://t.co/B7WzRetueM https://t.co/8JAs8mvkVH
— Rohan Paul (@rohanpaul_ai) Feb 5, 2025
OpenAI Sales Associate Agent spotted in recent Tokyo talk (zoom in, pic is from livestream) https://t.co/vVIHLljQaY
— Chubby♨️ (@kimmonismus) Feb 5, 2025
Stanford's "Deep Generative Models" 📽️ Lecture Videos: https://t.co/PzzajhvmRo 🗒️ Course Website: https://t.co/KbFDHbn3X4 https://t.co/KRfWMFrA3g
— Math Cafe (@Riazi_Cafe_en) Feb 4, 2025
This chart is insane. AI has now enabled the creation of the fastest growing software product maybe of all time. https://t.co/60i31j9olm
— Aaron Levie (@levie) Feb 5, 2025
NEW: Google just released Gemini 2.0 Pro experimental — the company's most advanced AI model yet. 2.0 Pro Experimental features include: — a massive 2M token context window — the ability to use tools like Google Search and code execution — Upgrades in reasoning, coding, and… https://t.co/uTVvgTb6Qw https://t.co/4TldPINOWf
— Rowan Cheung (@rowancheung) Feb 5, 2025
Within 24 hours, OpenAI's Deep Research has been replicated by an open-source version that already scores 54% on the same validation set OpenAI's scored 67%. https://t.co/uIUmaDjqcD
— Chubby♨️ (@kimmonismus) Feb 5, 2025
wake up, chat! @rasbt dropped reasoning in LLMs blog https://t.co/lzQfPHWMyf
— himanshu dubey (@himanshustwts) Feb 5, 2025
This guy literally drops a free masterclass on AI, robotics and computing. Can’t believe it’s free. https://t.co/KsqeLk2QKd
— Aadit Sheth (@aaditsh) Feb 4, 2025
a new incredible AI tool just dropped. now you can DESIGN lifelike voices with precise emotion control and.. even clone any voice with just a 10s clip step by step tutorial: https://t.co/YMYJzKfq0Z
— el.cine (@EHuanglu) Feb 3, 2025
Looking for better ways to code with Cursor and came across this banger. Anyone else have workflows that work for them? https://t.co/ZdDaabCCdD
— Alex Reibman 🖇️ (@AlexReibman) Feb 4, 2025
Wow, someone just released a notebook to train a reasoning LLM with the new RL algorithm from DeepSeek, GRPO. In <2 hours, you can transform a very small model, Qwen 0.5 (500 million parameters) into a tiny math reasoning machine. https://t.co/Su0cJ6kw9H
— Lior⚡ (@LiorOnAI) Feb 4, 2025
Excited to introduce Icon, The First AI Admaker. We’re backed by Peter Thiel’s Founders Fund & execs of frontier AI labs like OpenAI, Pika, & Cognition. Icon (https://t.co/Ze3NCCSld6) is like ChatGPT + CapCut, but for making winning ads with AI in minutes. How it works: 1.… https://t.co/p4N7XKxwlF https://t.co/2oI6Ctr8mE
— Kennan Davison (@kennandavison) Feb 4, 2025
China is on 🔥 ByteDance drops another banger AI paper! OmniHuman-1 can generate realistic human videos at any aspect ratio and body proportion using just a single image and audio. This is the best i have seen so far. 10 incredible examples and the research paper Link👇 https://t.co/FesUjaVEaS https://t.co/5OjNj0797t
— AshutoshShrivastava (@ai_for_success) Feb 4, 2025
Deep Research just dropped a wild list... 20 jobs that OpenAI o3 will replace humans. https://t.co/rB45IUqheC
— Min Choi (@minchoi) Feb 4, 2025
🤖🏭 Enterprise Agentic RAG Solution A new enterprise framework combines vector search and LangGraph agent orchestration for intelligent RAG applications. Delivers optimized retrieval with BBQ technology for 20-30x faster performance. Learn more about this enterprise AI… https://t.co/uImbrYF1wt https://t.co/N5e5TGfyZT
— LangChain (@LangChainAI) Feb 3, 2025
📈 Maximize Your Website’s Ranking Potential!https://t.co/ZPY2tp1fMo Free Account + 100 Free credits. Built with Love by @dharmesh
— Agent.ai (@AgentDotAi) Dec 2, 2024
