Our honest analysis of MIT Nanda's State of AI in Business report Read with your eyes open, and you'll see much more beyond that flashy 95% figure of organizations getting zero return from AI investments ⬇️ Find a full breakdown on YouTube: https://t.co/hdJq8pBIGP https://t.co/MkpgcrEJRp
— TuringPost (@TheTuringPost) Aug 30, 2025
gpt-5 is incredible at coding, and really shines with the right prompting style: https://t.co/BkdaM69Lln
— Greg Brockman (@gdb) Aug 30, 2025
OpenAI has released its "Realtime Prompting Guide". This is a paradigm shift in how to build an agentic AI system. Let's dig into the details of how different this is! https://t.co/vGYlPK5m4I
— Carlos E. Perez (@IntuitMachine) Aug 29, 2025
The data scientist slid the tablet across the table to the chief oncologist. On the screen was a single, stark prediction: "High Risk." The AI model had flagged a patient's scan for a rare, aggressive cancer. The oncologist looked up. "Why?" The data scientist tapped the https://t.co/PFQZk0XcJX
— Carlos E. Perez (@IntuitMachine) Aug 30, 2025
If something could kill Jupyter notebooks, this would be it. Zerve is a web-based IDE built for data scientists from the ground up. It's block-based (like a notebook), but the experience is very different. This alone is worth trying, but what sets Zerve apart is its Data https://t.co/l5o1530P8q
— Santiago (@svpino) Aug 29, 2025
Introducing ⚡️Cerebras MCP Server ⚡️ You can now turbocharge any AI editor that uses MCP for tool calling with 20x faster inference. Here is Claude Code writing files at breakneck speed via Cerebras. Get started now 👇 https://t.co/Mv206wgeb2
— Cerebras (@CerebrasSystems) Aug 28, 2025
Parallel agents are emerging as an important new direction for scaling up AI. AI capabilities have scaled with more training data, training-time compute, and test-time compute. Having multiple agents run in parallel is growing as a technique to further scale and improve
— Andrew Ng (@AndrewYNg) Aug 28, 2025
I asked the AI to plan a project budget. It was beautiful. Spreadsheets, timelines, cost breakdowns. I presented it to my boss. She pointed to one cell. A single, tiny formula error in step 2. It made the entire final number wrong. Catastrophically wrong. The AI failed. Not https://t.co/I5bmjCuJFJ
— Carlos E. Perez (@IntuitMachine) Aug 28, 2025
Microsoft's solution to roadmap planning when things change monthly/weekly/daily/hourly: Planning in "seasons" https://t.co/Izcjt4oyDE https://t.co/HHmF99reD1
— Lenny Rachitsky (@lennysan) Aug 28, 2025
Computer use is the next step towards true agentic coworkers. Models that can click, type, and reason across the existing software humans use will work like magic. Computer-using agents will actually provide end-to-end automation across legacy and modern tools alike: navigating https://t.co/86kwID8pKF
— a16z (@a16z) Aug 28, 2025
Palantir's Forward Deployed Billionaires have created over $44B+ in value founding companies like.. FDE → Founders: Chapter ($1.5B) - Cobi Blumenfeld-Gantz Kalshi ($2B) - Tarek Mansour Sourcegraph ($2.6B) - Quinn Slack Ironclad ($3.2B) - Cai Wangwilt Addepar ($3.25B) - Joe https://t.co/YPQmyOOEyt https://t.co/aPIzjSL6i7
— Molly O’Shea (@MollySOShea) Aug 29, 2025
Main points from $ANET @AristaNetworks at Deutsche Bank Tech conference on AI networking and AI demand. The age of giga-clusters is upon us. https://t.co/f1F7FhJVdp
— Ben Bajarin (@BenBajarin) Aug 28, 2025
Microsoft just dropped MCP with Python course for FREE! This 3 part course starts from basics and helps you build your first MCP server. Find the course details 🧵👇 https://t.co/6LyzuoDPev
— Afiz ⚡️ (@itsafiz) Aug 27, 2025
AI Evals are critical for AI PMs and Engineers. But many still confuse them with unit tests. By popular request, I removed a paywall from the @HamelHusain's Mastering AI Evals, A Complete Guide. Key insights, free guide, and massive resources: 🧵 https://t.co/nSKSNr6LWC
— Paweł Huryn (@PawelHuryn) Aug 27, 2025
So many crazy applications that wouldn't work just a few days ago but now do work with Nano Banana Upload any flat lay "get the look" type pic Prompt: "show woman wearing the outfit" It outputs highly accurate pics with a woman wearing the exact outfit It's really a wow https://t.co/bfRowY5Q2S https://t.co/FrhDbjqwqk
— @levelsio (@levelsio) Aug 28, 2025
BREAKING: Unpacking the largest investment in the history of Altimeter → OpenAI Reaching a $500B valuation & 700M WAUs, Apoorv Agrawal, AI & Software Partner at Altimeter, breaks down how to value OpenAI's insane growth We dig into the numbers. + PLTR FDE, XBOW, Power Laws, https://t.co/rtZFcksT8B
— Molly O’Shea (@MollySOShea) Aug 27, 2025
I just listened to a breakdown of Gemini jump from 2.0 to 2.5, and here's the gist. The team did not rely on lab tests alone. They scraped real user feedback from X, turned those "this broke" moments into living evals, and keep appending new edge cases with every release. It's https://t.co/qRzgOVvHRO
— Matija Grcic (@matijagrcic) Aug 27, 2025
NEW: Assort Health, an AI voice agent startup for specialty care, raises $50M Series B at a $750M valuation Lightspeed led the round, just 4 months after Assort’s $22M Series A. The startup automates patient scheduling and communication with AI-powered phone agents. https://t.co/CXbkLFOEtX
— Healthcare AI Guy (@HealthcareAIGuy) Aug 27, 2025
SuperGrok is insanely powerful. Use the file chat feature and talk to your code in real time. Coded this 3D model of helmet for a game asset using threejs and webgl with Grok 4. The file chat allowed me to add lighting and color controls just like that. It felt surreal. Try it https://t.co/aFHcfpIqxL
— Prashant (@Prashant_1722) Aug 26, 2025
The only prompt you need to be a Prompt Engineer! https://t.co/RrHUtHRfDp
— Pradeep Pandey (@Div_pradeep) Aug 26, 2025
📁 Anton Osika, co-founder and CEO of Lovable, says that: “We’re entering a new era where software won’t be written in code, but created by talking to AI.” “Today, less than 1% of the world can code, soon, the other 99% will finally be able to build.” “The barrier to creating https://t.co/6NVyede0dz
— Jon Hernandez (@JonhernandezIA) Aug 25, 2025
Using https://t.co/DNJ20jdplL to spin up content strategy docs based on top Youtube videos on the fly. Rube brings the Tab Tab Tab experience outside of Cursor / Claude to the real world. https://t.co/FaNohhkS6T
— Sujay (@sujaychoubey) Aug 25, 2025
The Lonely Chapter of Reinvention. https://t.co/4DEqrPPH5r
— Chris Williamson (@ChrisWillx) Aug 25, 2025
Just tried 8090 Factory Alpha (cc @chamath). It slows you down to speed you up — forcing clarity, specs, and discrete units of work instead of messy one-shot builds. Check out my review... https://t.co/pMA2X3v6Xv
— Chris Stanchak (@chrisstanchak) Aug 25, 2025
Masterclass https://t.co/rj7IjAcc3S
— Garry Tan (@garrytan) Aug 26, 2025
I've been waiting for this. The Unified MCP is here! Rube is a universal MCP server to connect your AI agents to all *your* apps. Works with your favorite IDE, Claude Code, and other MCP clients. Watch it research YouTube vids → generate a full content strategy doc. Insane! https://t.co/KRlhIjsFJ0
— elvis (@omarsar0) Aug 25, 2025
What exactly are AI agents, and how will they change the way we work? @levie, @martin_casado, @stevesi, @eriktorenberg discuss: (0:36) What Is an Agent? Agency, Autonomy, and Feedback Loops (4:27) Task Specialization and Division of Labor in AI (6:04) Anthropomorphizing AI https://t.co/QfoNZF8kEZ
— a16z (@a16z) Aug 25, 2025
This is from my conversation with a Former $NVDA Director. If you want more TOKENS per SECOND because you are limited by POWER, you buy the next $NVDA generation. - Limited to 100 MW: $NVDA H100 = 300M tokens/second - Limited to 100 MW: $NVDA GB200 NVL72 = 12B tokens/second https://t.co/pHCp1JQIh8
— Rihard Jarc (@RihardJarc) Aug 25, 2025
List of all of the n8n workflows at one place. https://t.co/YzXOYLETHE
— GitHub Projects Community (@GithubProjects) Aug 25, 2025
If we have 100X more AI agents than people doing work in a company, it’s actually much more likely that we conform to how agents work best instead of agents conforming to how we work. AI agents fundamentally thrive on context, and the variance in what you get out of them almost
— Aaron Levie (@levie) Aug 25, 2025
🔋A very insightful 26-page report from Goldman Sachs on Energy and AI. Powering the AI Era Says a fresh power‑grid crunch on the horizon because AI server farms soak up electricity faster than utilities can add capacity, and it thinks creative financing, not just faster chips, https://t.co/eLazxcPCrr
— Rohan Paul (@rohanpaul_ai) Aug 24, 2025
// From AI for Science to Agentic Science // New survey paper on autonomous scientific discovery. It covers open challenges in reproducibility, novelty validation, and transparency. It outlines future directions like global cooperative research agents and a Nobel‑Turing Test. https://t.co/s7IcjbBNNO
— elvis (@omarsar0) Aug 24, 2025
Meet AETHRA-E...a drone without rotors or wings. Slot microjets embedded in the hull vector a continuous thrust sheet to lift, steer, and stabilize. Edge-blown monocoque: the airframe itself becomes the nozzle. Just shaped thrust. No props. No wings. https://t.co/hPuTdmpH2J
— Mathelirium (@mathelirium) Aug 24, 2025
More from the SSTorytime project ... how to avoid common mistakes when designing knowledge graphs https://t.co/jjKxFwJJS7
— Mark Burgess (@markburgess_osl) Aug 24, 2025
Here's a production-ready pattern for adding 'suggested next questions' to any AI app with the AI SDK. Want to see inside these functions? Check out the thread below. https://t.co/EBaN5OBqVc
— Matt Pocock (@mattpocockuk) Aug 24, 2025
BREAKING 🔥 @xai just revealed Macrohard Doors — the first AI-only operating system. Introducing nOS: a neural operating system where nothing is hardcoded, everything is created on the fly by @grok. Here’s your first look at the OS’s visual style: https://t.co/K9olPbuKsg https://t.co/9c1CZ2Cugd
— tuōmo (@7uomoki) Aug 24, 2025
The debate around whether every pixel in a photo from your phone's camera is "real" misses a fundamental fact about how digital cameras have always worked for the last 20 years. The camera sensor only captures ONE color (red, green, or blue) per pixel. The rest are made up 1/4 https://t.co/zBNVW2034D
— Peyman Milanfar (@docmilanfar) Aug 23, 2025
best tool roundup I’ve seen in a while.. and you can check https://t.co/HxI61DUNTH for custom agents on any use case https://t.co/O1PS36JyxR https://t.co/R8UtgvVZgb
— Nelly; (@nrqa__) Aug 24, 2025
Founder mode vs manager mode made visible in logos. Cracker Barrel is just the latest instance. https://t.co/xWsHluycuK
— Paul Graham (@paulg) Aug 23, 2025
We hit $10M ARR in 2 months making us one of the fastest growing AI startups in the world 🚀 And today, we are putting more power into the hands of people while changing the world of vibecoding 🙌 Introducing: 1. Emergent Pro Mode- Build real autonomous agents 2. Mobile Apps- https://t.co/rDBgPNFS6r
— Emergent (@EmergentLabsHQ) Aug 23, 2025
If you’re building AI systems in 2025, there are only two tools worth learning: LangGraph and n8n. The choice you make here will define how far you can actually scale. Here’s everything you need to know (and what nobody is telling you): https://t.co/gbHYhnh0WD
— Connor Davis (@connordavis_ai) Aug 23, 2025
n8n is everywhere now... my friend's construction company uses it, accounting firms are adopting it and even restaurants are running workflows this creates a paradox: - easier to sell automation services because businesses understand the value - but harder to compete because
— Machina (@EXM7777) Aug 22, 2025
Introducing NEO: The first Autonomous Machine Learning Engineer. It works like a full-stack ML engineer that never sleeps: handling data exploration, feature engineering, training, tuning, deployment, and monitoring, end to end. Powered by 11 specialized agents, NEO runs https://t.co/7MpnESSBMG
— Neo AI (@withneo) Aug 22, 2025
LFG https://t.co/6wwlaqM95r https://t.co/FR0WEQYycS
— Lenny Rachitsky (@lennysan) Aug 22, 2025
Databricks plans to acquire Tecton, which helps companies build and deploy ML applications, for an undisclosed sum; Tecton was valued at $900M in 2022 (@readkrystalhu / Reuters) https://t.co/CMCE61JEPb https://t.co/OtdCdFZ5nE
— Techmeme (@Techmeme) Aug 22, 2025
Just attended @every 's Claude Code Camp and wow 🤯 Loved seeing how @DannyAziz97 uses one agent to build UI from Figma while another agent critiques it until they get it right. Kudos to @danshipper, @kieranklaassen, @poojary_yash for sharing their learnings with community of https://t.co/y4JTg1OsEZ
— Amit Bhatia (@ridegoodwaves) Aug 22, 2025
Everyone is posting about the “95%” Fortune article, but it’s worth reading the underlying MIT report, which contains a ton of interesting insights. My main takeaway is that Gen AI projects fail not because the models aren’t capable, but because the importance of context is https://t.co/JVRQ7CTqBb
— Douwe Kiela (@douwekiela) Aug 21, 2025
Anthropic slayed in this 1-pager about how to use AI at work https://t.co/krMeLF35um
— Aakash Gupta (@aakashg0) Aug 21, 2025
AI Workflow Details https://t.co/rgJjpDQjbZ
— Kevin (@kvnkld) Aug 21, 2025
Google just dropped a paper showing that the energy used for a Gemini text prompt has dropped 33x in only 12 months btw https://t.co/gAZRPPtUmB https://t.co/ebYiMDx6dB
— NIK (@ns123abc) Aug 21, 2025
Remember Matryoshka embeddings? They're still incredibly useful, but it feels like everyone's forgotten about them. We're all trying to optimize for cost and speed in our AI apps, yet many developers are still using full-size embeddings when they could get 90%+ of the https://t.co/RwSZEl9tZe
— Victoria Slocum (@victorialslocum) Aug 21, 2025
$PLTR 2 Patents https://t.co/REyGMSan0I
— Gokarp (@PLTRs_Palantir) Aug 21, 2025
Using @elevenlabsio v3 with lip-sync models is such a strong use case 🤯 perfect for creating natural talking avatars. You can try it out on @FAL ! https://t.co/CyS30C46sE https://t.co/757DkgFUIl
— İlker (@ilkerigz) Aug 20, 2025
An expert wants to know things with certainty and in the end to be proven right. Approaching life with a beginner’s mindset frees you from the past - attachments, fears, preconceptions even aspirations - to an open mind and heart in the present moment. —@Benioff TRAILBLAZER https://t.co/zkrjKcErF1
— Vala Afshar (@ValaAfshar) Aug 20, 2025
Peter Thiel on what it takes to build monopoly-style businesses. https://t.co/4EbzMIIm9a
— Z Fellows (@zfellows) Aug 20, 2025
RAG is dead?? HA, you just haven’t tried agentic RAG yet… Announcing Elysia - an open source LLM agentic Python package that doesn’t just call tools. Featuring both a 𝗶𝗻-𝗯𝘂𝗶𝗹𝘁 𝗳𝗿𝗼𝗻𝘁𝗲𝗻𝗱 𝘄𝗲𝗯 𝗮𝗽𝗽 and an 𝗲𝗻𝘁𝗶𝗿𝗲 𝗣𝘆𝘁𝗵𝗼𝗻 𝗹𝗶𝗯𝗿𝗮𝗿𝘆, Elysia is https://t.co/jLCAlRxyZC
— Victoria Slocum (@victorialslocum) Aug 20, 2025
In building AI agents @cline , we've identified three mind viruses Mind Viruses are seductive ideas that sound smart, but don’t work in practice. 1. Multi-Agent Orchestration 2. RAG (Retrieval Augmented Generation) 3. More Instructions = Better Results Let's explore why! https://t.co/p2p93Dt9QM
— Ara (@arafatkatze) Aug 19, 2025
This is actually a really solid context engineering template. Kudos, @AnthropicAI https://t.co/hKuxNK3h48
— Matt Pocock (@mattpocockuk) Aug 20, 2025
Have you read Camus and felt like he’s speaking directly to your soul? https://t.co/rubRCe4vTa
— Friedrich Nietzsche Notebook (@QuoteNietzsche) Aug 19, 2025
i love that we have our own little universe of words to talk about the future @every: allocation economy compounding engineering model manager multidimensional knowledge orchestration alien superintelligence ai operations
— Dan Shipper 📧 (@danshipper) Aug 19, 2025
𝗔𝗜 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗥𝗼𝗮𝗱𝗺𝗮𝗽. 👇 It is created with beginners in mind but can be easily adapted if you are proficient in some of the areas already. 𝘖𝘯 𝘢 𝘩𝘪𝘨𝘩 𝘭𝘦𝘷𝘦𝘭: 𝗙𝗼𝗰𝘂𝘀 𝗼𝗻 𝗙𝘂𝗻𝗱𝗮𝗺𝗲𝗻𝘁𝗮𝗹𝘀 throughout the journey, but https://t.co/Sc4h1QeYiB
— Aurimas Griciūnas (@Aurimas_Gr) Aug 19, 2025
I don’t know if this actually is the future (bc it’s unclear to me to what extent we can actually meaningfully improve LLM usage) but I think we have to try, and this is a good framework for trying. https://t.co/6iHBUL4yFQ
— Nate Berkopec (@nateberkopec) Aug 18, 2025
🎯 https://t.co/FvOax1Qpsl
— Eric Buess (@EricBuess) Aug 18, 2025
Congrats to Anna and the Paradigm team on their launch. I’ve been using Paradigm every week to save me hours of time. Try it out! https://t.co/sh1ixiYmBA
— Michelle Lim (@michlimlim) Aug 18, 2025
Imagine learning the latest research/practice on MoEs, quantization, positional encodings, and other foundation model topics in single week by listening to this mixture of wonderful experts. Don’t miss it! https://t.co/2Q9koUK4a4
— Omar Khattab (@lateinteraction) Aug 18, 2025
Wow, BlackRock just proposes AlphaAgents for investment research! 🤖💵💴💶💷 Equity portfolio management has long relied on human analysts poring over 10-Ks, earnings calls, and market data—a process that’s slow and prone to biases. Enter AI-powered multi-agent LLMs: teams of https://t.co/uOJQt1pUbw
— 机器之心 JIQIZHIXIN (@jiqizhixin) Aug 18, 2025
The Knight --sref 3550802132 https://t.co/cffhYdVny8
— Ege (@egeberkina) Aug 18, 2025
WTF… While GPT‑5 disappointed everyone.. China quietly dropped the most advanced AI agent system we’ve seen yet. – Fully autonomous – 100+ agents in parallel – Interrupt tasks mid-run 🔥 – Real-time memory + reasoning Silicon Valley can't afford to ignore this: 🧵 https://t.co/SSU3PPC3XU
— EyeingAI (@EyeingAI) Aug 19, 2025
Learn to fine-tune OpenAI gpt-oss with our new step-by-step guide! Learn about: • Local gpt-oss training + inference FAQ & tips • Evaluation, hyperparameters & overfitting • Reasoning effort, Data prep • Run & saving your LLM to llama.cpp GGUF, HF 🔗https://t.co/Vpd72FdEw3 https://t.co/jq9XqZNrTY
— Unsloth AI (@UnslothAI) Aug 18, 2025
The future looks like it may just be a billion RL environments https://t.co/EatuEu1nPQ
— Teknium (e/λ) (@Teknium1) Aug 18, 2025
This AI agent just watched 247 TikTok videos & extracted every winning hook in under 5 minutes 🤯 Most ecomm brands & agencies spend WEEKS manually scrolling TikTok for creative inspiration. But this n8n + Airtable system does it automatically: → Enter any keyword (hair care, https://t.co/t4qtegVbol
— Mike Futia (@mikefutia) Aug 17, 2025
"The doers are the major thinkers. The people that really create the things that change this industry are both the thinker/doer in one person. Leonardo was an artist but he also mixed his own paint. He was also a fairly good chemist, knew about pigments." — Steve Jobs, 1990
— Steve Jobs Stories (@SJobs_Stories) Aug 17, 2025
The paradigm of AI subagents is going to be super interesting. There was probably some hope or belief that a universal agent would be able to handle everything you needed in a workflow by stuffing all the relevant context into the context window. But even with larger context
— Aaron Levie (@levie) Aug 18, 2025
Shower thought: in my lifetime I’ll have witnessed the very bottom of the deepest bear market in Christianity—250 years in the making—and the beginning of its revival. https://t.co/lUs8D3l1aT
— Tuur Demeester (@TuurDemeester) Dec 29, 2024
"Scene description" is more than describing a scene. It's a means by which you express YOUR WRITER'S VOICE! Learn tools to convey your unique voice on the page. One-week online class starts Monday. https://t.co/T5io0f8Xc9 #screenwriting https://t.co/CJc4kgZkeo
— Scott Myers (@GoIntoTheStory) Aug 17, 2025
Building compound AI systems is the new data visualization. It shows up in Excel, Python, and R. IDEs and tools are now built specifically to support it well (Tableau, RStudio, or Jupyter Notebooks). There are books and courses about it, and it’s taught to accountants,
— Maxime Rivest 🧙♂️🦙🐧 (@MaximeRivest) Aug 17, 2025
The open-source desktop agent devs have been waiting for. Your AI deserves more than a browser tab. https://t.co/ML9llfnIpo
— GitHub Projects Community (@GithubProjects) Aug 16, 2025
LinkedIn Recruiter AI search is one of the best AI search products I've ever used. Exa Websets sometimes gets stellar hits, but is far less comprehensive/thorough. https://t.co/TZYzrz4CXd
— Charlie Wu (@ccharliewu) Aug 17, 2025
🌪️Prompt Share: Veo 3 text to video A colossal ancient statue emerging from desert sands during a windstorm; low-angle crane shot, warm sepia haze, dramatic reveal. https://t.co/vFz9iA0hHF
— Heather Cooper (@HBCoop_) Aug 16, 2025
I am (slowly) re-reading the Tolkien legendarium (of which Lord of the Rings is a small part). The whole body of work is so incredible and there's nothing else like it... it dilutes other worlds of fiction. Wait - your story doesn't have a comprehensive history/mythology spanning
— Andrej Karpathy (@karpathy) Aug 16, 2025
Steve Jobs dropped the coldest business truth of all time https://t.co/7r7JSqubcS
— Aadit Sheth (@aaditsh) Aug 15, 2025
i don't think this company @every is being talked about enough... they are making some really interesting apps/services AND they also do media wildly underrated combo https://t.co/PkL2v4up89
— Bo (@bonam) Aug 15, 2025
Grok Imagine just brought this moment to life. Elon Musk in 2006, staring at debris from Falcon 1, which was the first of three failed rocket launches from SpaceX’s early days. https://t.co/xqYzHfSP8q
— SMX 🇺🇸 (@iam_smx) Aug 15, 2025
Sam Altman speaks on GPT-5: > we had this big GPU crunch > we could’ve made another giant model > people would want it… and we’d disappoint them lol > so we made a *really smart, really useful* model instead > optimized for inference cost LMAO https://t.co/A3TSnBqkUi
— NIK (@ns123abc) Aug 15, 2025
Very thoughtful, wide-ranging interview with ChatGPT product lead @nickaturley. https://t.co/OMaOkFjyCB https://t.co/te5y6eQOyN
— Jeff Weinstein (@jeff_weinstein) Aug 15, 2025
One subtle reason why the market size for AI Agents is going to be bigger than we realize is there are many use-cases where running multiple agents in parallel to solve the same problem is valuable. This was effectively never possible with most knowledge work before. It just
— Aaron Levie (@levie) Aug 16, 2025
https://t.co/Zg8Xptl6ZD never sleeps https://t.co/GNtqy581MA
— Lenny Rachitsky (@lennysan) Aug 14, 2025
Claude 4.1 Opus System prompts have been leaked! Let's delve into the details of how it's constructed. The Claude 4.1 system prompt is a highly structured and modular configuration that exhibits several recurring system design patterns described in the Pattern Language for https://t.co/BxOeffS7to
— Carlos E. Perez (@IntuitMachine) Aug 14, 2025
Google DeepMind just dropped a Nature paper: “A personal health large language model for sleep and fitness coaching.” > Gemini is better that doctors and trainers at sleep and fitness > paper shows huge benefit of AI personalization for health and long-form coaching Honestly https://t.co/2e2zsFOEBe
— NIK (@ns123abc) Aug 14, 2025
Much to learn. https://t.co/0hvnpYTvDL
— Suhail (@Suhail) Aug 14, 2025
First of the few changes from Booster Block 2 to Block 3 that is being acknowledged by SpaceX that was shown but not touched upon during @elonmusk's presentation in May. The new information is that the Booster will now be caught using the grid-fins rather than the tiny lifting https://t.co/RlU6m8IYRn https://t.co/m2axNBiPUu
— Vikranth (@VikranthJonna) Aug 14, 2025
Founders meet at @southpkcommons -> S24 Founder Fellowship -> Series B from @sequoia in just over a year Velocity here is crazy, but unsurprising if you know the @tryprofound team https://t.co/yxV3sdqDxb
— Jonathan Brebner (@JPBrebner) Aug 12, 2025
Traditional RAG vs. Agentic RAG, visually explained: https://t.co/sg3RHn8GHF https://t.co/iBtnm27rmv
— Akshay 🚀 (@akshay_pachaar) Aug 13, 2025
Customers don't Google blue links to decide what to buy anymore. They ask ChatGPT. Google currently drives $2.4 trillion of commerce. In 5 years, $1 trillion of that will shift to ChatGPT. Becoming ChatGPT's #1 recommendation could be worth $100M+ to your business. Here's what https://t.co/T6UFyz2CrT
— James Cadwallader (@thejamescad) Aug 12, 2025
Naval is right. That’s why we’ve avoided investing in consumer hardware. But recently I’ve met too many smart, young founders building it anyway that it’s too compelling to ignore. We’ve built an AI god and shoved it in a text box. Like now what? That can’t be the end of the https://t.co/apUP4tvYUO https://t.co/Pgjsoes6d5
— Bryan Kim (@kirbyman01) Aug 13, 2025
Generative vs. discriminative models in ML: (a popular ML interview question) https://t.co/bOWdx8CywA
— Avi Chawla (@_avichawla) Aug 13, 2025
This one is going to be great. https://t.co/cEw792srj7
— Patrick Collison (@patrickc) Aug 12, 2025
LangChain literally reverse-engineered Claude Code and Manus AI to build Deep Agents. It's a Python library to turn any LLM into a deep thinking agent with MCP tools. 100% opensource. https://t.co/Coput1w9QN
— Shubham Saboo (@Saboo_Shubham_) Aug 12, 2025
How do you build an end-to-end agentic RAG app? Lucky for you, you can just run two commands: ‘pip install elysia-ai’ and ’elysia start’ I wrote a massive blog post detailing all the things we built into this 𝗼𝗽𝗲𝗻 𝘀𝗼𝘂𝗿𝗰𝗲 𝗲𝗻𝗱-𝘁𝗼-𝗲𝗻𝗱 𝗮𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗔𝗚 https://t.co/Ybi9bzhifY
— Victoria Slocum (@victorialslocum) Aug 12, 2025
Kyoto, Japan. https://t.co/GdIV4ptWcv
— JAPAN 🇯🇵 (@Japantravelco) Aug 11, 2025
After nearly four years as CEO, I’m leaving GitHub to become a startup founder again. With more than 1B repos and forks, 150M+ developers, and Copilot continuing to lead the most thriving market in AI with 20M users and counting, GitHub has never been stronger than it is today.
— Thomas Dohmke (@ashtom) Aug 11, 2025
The GLM-4.5 technical report is out! Sharing some key details in case you missed it: https://t.co/cfg3DGxtjf
— elvis (@omarsar0) Aug 11, 2025
Co-creator of the $PLTR logo @garrytan on what Founders should learn from Forward-Deployed Engineers 🔮🥩 “You can’t farm this out” “It’s no mistake that FDEs who came from that system at Palantir now are turning out to be some of the best founders at @ycombinator” “Instead of https://t.co/xOhMcGdYe5
— Jawwwn (@jawwwn_) Jun 2, 2025
This AI TikTok page made $1,645,789 in 102 days. No dices. No cooking. No chef. Just: - 300+ AI recipe videos/month - Faceless voiceovers + AI food images - $10 cookbook sold on autopilot - 35M+ views | Dozens of 1M+ videos It looks like a real dices — but it’s 100% AI. Want https://t.co/lwyshSy0uo
— Cas.Fyn (@FynCas) Aug 10, 2025
👨🔧 Github: This Github literally has hundreds of System Prompts 74.5K stars 🌟 v0, Cursor, Manus, Same .dev, Lovable, Devin, Replit Agent, Windsurf Agent, VSCode Agent, Dia Browser, Xcode, Trae AI, Cluely & Orchids. app --- github. https://t.co/FYKRQy6PLf
— Rohan Paul (@rohanpaul_ai) Aug 10, 2025
GPT-5 as a scientific collaborator: https://t.co/AWSRjOdmLu
— Greg Brockman (@gdb) Aug 10, 2025
A must watch video for all TSLA investors. Adam Jonas does an excellent job of interpreting the future of real world AI. Spoiler alert, manufacturing is key. https://t.co/dZxEkiAb8F
— Simon_Hale 🇨🇦 (@SimonHale007) Aug 10, 2025
Claude Code as your Devops agent. The new background jobs (like cursors shared terminals) make this way easier. 1. start claude in tmux session 2. ask it to create background process to tail your server logs and give you a summary 3. then ask it to run a second process that https://t.co/VqBvi3XF90
— Imrat (@imrat) Aug 10, 2025
Grok is one-shot coding increasingly complex video games. Our next update will actually *play* the game, look at the screen and adjust the code to improve both esthetics and playability. https://t.co/3ZVWv4QwJ7
— Elon Musk (@elonmusk) Aug 10, 2025
N8N FULL COURSE 5 HOURS (Build & Automate Anything) https://t.co/PgIQGchqlL
— Julian Goldie SEO (@JulianGoldieSEO) Aug 10, 2025
Most are still sleeping on this: the choice of program structure (how modules are composed, use of retrieval, ranking, fusion, etc.) is important to both quality and efficiency. How you compose & build your AI system creates real value. It is a moat, and creates meaningful IP. https://t.co/A34LANBYqU
— Gabriel Lespérance (@GabLesperance) Aug 9, 2025
Steve Jobs drew heavily on Japanese philosophies to shape his leadership style. Thanks to my good friend @RepaVictoria for reminding me of something powerful: Here are 5 Japanese techniques every leader should master 👇 ----- 📌 Investment MBA is launching soon: The top https://t.co/v0OjXkJQor
— Igor Buinevici (@Igor_Buinevici) Aug 8, 2025
🇨🇳 A BATTERY THAT LASTS LONGER THAN YOUR FAMILY TREE China’s Betavolt BV100 is a coin-sized nuclear battery that can power stuff for over 50 years without ever needing a recharge. It uses nickel 63 atoms and diamond semiconductors to turn radioactive decay into safe, steady https://t.co/mOvGe5ymn2
— Mario Nawfal (@MarioNawfal) Aug 9, 2025
“More designers need to be founders”, and @zoink is leading the way. Before the @figma IPO, Dylan came to AI Startup School to share the Figma founding story, why great design is more important than ever, and how we’re still in the MS-DOS era of AI. https://t.co/iygBlSbOFp
— Aaron Epstein (@aaron_epstein) Aug 9, 2025
Harmony: the new prompting standard? With the release of open weigh GPT-OSS, @OpenAI proposed a solution for prompt templating: ▪️ Harmony - an open-source (Apache 2.0) response format mirroring OpenAI's proprietary Responses API. What's new for the open-weight world? • https://t.co/gmj5AyQwkZ
— TuringPost (@TheTuringPost) Aug 9, 2025
Keep swinging. Bezos viewed each experiment as a call option: limited downside, massive potential upside. One of those experiments, AWS, alone turned into a $100b biz line now (and eventual trillion-dollar business line), more than covering the “cost” of every failed project https://t.co/mhPhDQnxOW
— adam bain (@adambain) Aug 9, 2025
Every time I see it, I think — it’s incredible that we can make this. Prompt: Two anime characters fight in a blank white void — with every hit, their surroundings are drawn into existence by the motion — buildings, trees, skies appear stroke by stroke as the fight intensifies https://t.co/WMOBwCpRM6
— OscarAI (@Artedeingenio) Aug 7, 2025
To the Ivory Tower AI Doomers, AI is anti-human and job destructive. That's their choice. Our experience is AI makes human taste and talent exponentially more valuable. We've watched AI unleash human agency and creativity. That's how you give the American worker https://t.co/DPzT7Zw38i
— Shyam Sankar (@ssankar) Aug 7, 2025
This Midjourney Sref is SOOO GOOOD! You must use this! --sref 4053907766 https://t.co/k22OYW7lIo
— Keskin (@craftian_keskin) Aug 7, 2025
@thechangj @ben_aguilera is the king at this stuff 👑
— VCs Congratulating Themselves 👏👏👏 (@VCBrags) Aug 8, 2025
worked with @thenicktrimmer and @Pri_promo on our new launch video. Most talented storytellers/product marketers/video editors I've met. I'm genuinely envious of their skill. ps you'll see the video on Aug 19th so u can judge for yourself
— Dimi Nikolaou (@dimireadsthings) Aug 2, 2025
2025.32: What Nokia Can Teach Us About the AI Era The best Stratechery content from the week of August 4, 2025 including what Nokia can teach us about the AI era, what the NFL wants from ESPN, and how Visa conquered debit cards. https://t.co/10c8Qg2wmZ
— Stratechery (@stratechery) Aug 8, 2025
Here's BlockDL, a free & open-source GUI that lets you visually design Keras neural networks and learn ML. https://t.co/3r4qz02ghZ https://t.co/iTR2h4c8et
— François Chollet (@fchollet) Jul 29, 2025
Finally, a 45 page literature review of text embedding model, datasets, evaluation and training methods: https://t.co/EmVNNFR7KA
— Bo (@bo_wangbo) Jul 29, 2025
https://t.co/Fp9IqhzXRE
— Jack Prescott (@JackPrescottX) Aug 7, 2025
Proud to announce @ChaiDiscovery has raised a $70M Series A from @MenloVentures & Anthology Fund (@AnthropicAI), @ThriveCapital, @OpenAI + others. 🧬 Chai is an applied AI lab building frontier tools for molecular design. This funding will power our new models, coming soon👇 https://t.co/HdMPIcCVC1
— Joshua Meier (@joshim5) Aug 6, 2025
Synthetic Society tests your product with AI-powered user simulations. Their agents mimic real users to catch bugs, bad UX, and edge cases. Ship faster, kill manual testing, and build with confidence. https://t.co/51uOlq9JKe Congrats on the launch, @aaronchewbani and @kavandoc! https://t.co/pl1TwC11Fp
— Y Combinator (@ycombinator) Aug 6, 2025
the gpt-oss model is really easy to tune! get started with customizing/fine-tuning to make gpt-oss your own with the @openai + @huggingface cookbook 🤝 https://t.co/IUsYPJOz2c https://t.co/4pMgpm7pqc
— apolinario 🌐 (@multimodalart) Aug 5, 2025
Forward Deployed Engineering is one of the fastest-growing roles in tech 🔥 Here’s how we built and scaled Ramp’s FDE team to win in enterprise — and what we learned. 🧵👇
— Leo Mehr (@LeoMehr) Aug 5, 2025
This Google Veo 3 "frames to video" feature is pure genius 💡 If you're creating video ads, this Veo 3 technique gives you surgical control over every scene element. This frames-to-video feature will take any static product shot... And turn it into cinematic reels with precise https://t.co/smIbFwb5rW
— Mike Futia (@mikefutia) Aug 5, 2025
It's over. Pack it in. The slop is now A5 Wagyu. Netflix, Spotify, Game Designers, etc. will all be using this for background music. Why pay real composers when you can have an AI crank it out in minutes. The LoFi channels on YouTube just got infinitely generative music. https://t.co/MvUwoO5tga
— Autism Capital 🧩 (@AutismCapital) Aug 5, 2025
Introducing GPT OSS for understanding research papers 🚀 Highlight any section of a paper to ask questions and “@” other papers for quick context, comparisons, and benchmark references https://t.co/9BGeZspAb2
— alphaXiv (@askalphaxiv) Aug 5, 2025
Here's the concluding section of my write-up of the new OpenAI open weights models - they're really impressive, and I think they hold their own (or even beat) the models we've seen come out of the various Chinese AI labs over the past few months https://t.co/FqLuM3tEnY
— Simon Willison (@simonw) Aug 5, 2025
The new gpt-oss-120b from @OpenAI is now the best general purpose model to run on a Framework Desktop! ~40 tok/s on the MXFP4 version in @lmstudio on Fedora 42. https://t.co/YHWf9s6iDV
— Framework (@FrameworkPuter) Aug 5, 2025
This was from @ssankar 2012 Ted Talk: The rise of human-computer cooperation -10 years before ChatGPT Launch -11 Years before "Agents" were talked about 13 years later this thesis is proving out that Humans + Computers is the way and The Ontology is at the center of this https://t.co/CpbH9pOvxj
— Chad Wahlquist (@chadwahl) Aug 5, 2025
Anthropic has released a new prompt engineering guide and it’s one of the best out there. It covers meta prompting, templates, system roles, XML tags, and more. Here’s what it teaches (and how to level up your Claude prompts): https://t.co/R2Mg8pQqOS
— Rimsha Bhardwaj (@heyrimsha) Aug 5, 2025
Agents will become a common way people shop. So today we are releasing 3 tools to make adding commerce to those agents trivial: - Checkout Kit: embed commerce widgets and checkout(!) directly into your agent and chat. This is already being used by Microsoft’s @Copilot. - Shopify https://t.co/HSt6B8RtmF
— tobi lutke (@tobi) Aug 5, 2025
"The very best startup ideas tend to have three things in common: they’re something the founders themselves want, that they themselves can build, and that few others realize are worth doing." – Paul Graham https://t.co/GJDIkcsNL2
— Z Fellows (@zfellows) Aug 4, 2025
Steve Jobs explains the importance of both thinking and doing “The doers are the major thinkers. The people that really create the things that change this industry are both the thinker-doer in one person.” This is applicable outside of tech too, and he uses Leonardo DaVinci as https://t.co/hCUNF2Zcqc
— Startup Archive (@StartupArchive_) Aug 4, 2025
My biggest takeaway from this interview was the importance of GTM. Bret highlights how his startup, FriendFeed, got crushed by Twitter because they ONLY focused on product, not getting people on the platform. Twitter won because they onboarded celebrities to the platform https://t.co/x8VzwKvMWl
— Chase (@itsmechase) Aug 4, 2025
Napoleon: https://t.co/Lu4h4CUVGK
— David Senra (@FoundersPodcast) Aug 3, 2025
Richard Feynman's PhD thesis: "Principles of least action in quantum mechanics." https://t.co/J3xNmJzOHX
— Physics In History (@PhysInHistory) Aug 3, 2025
This is the wildest Grok 4 prompt I've ever used It can literally build you a business in one prompt Grok will research your audience, their challenges, then give you ideas for apps to build. If you take action on this post you'll have your own business BOOKMARK THIS Prompt:
— Alex Finn (@AlexFinnX) Aug 3, 2025
Using Claude Code just got 10x easier. Someone literally built 100+ agents, commands, MCPs, and templates that you can install and use with one command. And it's 100% free to use. https://t.co/Mk9GOyeZmS
— Shubham Saboo (@Saboo_Shubham_) Aug 3, 2025
The paper argues the next big thing is the Agentic Web where autonomous AI agents, not humans, run tasks across the internet. The authors show how the Web has already shifted from early pages to mobile feeds and now toward agent‑driven action. They define 3 dimensions, https://t.co/xxkojcEu3d
— Rohan Paul (@rohanpaul_ai) Aug 3, 2025
The Feynman path integral is a way of describing quantum mechanics using the concept of paths. A path is a possible way that a particle can move from one point to another in space and time. In classical mechanics, there is only one path that minimizes the action, which is a https://t.co/HlvkZLkXvZ
— Physics In History (@PhysInHistory) Aug 2, 2025
Genspark just dropped a Multi-Agent Platform for Vibe Working. One lead agent coordinates multiple specialized sub-agents to accomplish complex tasks simultaneously. No more juggling different tools - one Super Agent does it all! https://t.co/ELIfwPdrca
— Shubham Saboo (@Saboo_Shubham_) Aug 2, 2025
📹 Drone innovation at its peak: #drone which can run on wheels as well as fly with 8 different configurations #IADN https://t.co/dZM7t8yg11
— News IADN (@NewsIADN) Aug 2, 2025
