Gemini can now write Python code in Google Sheets You can generate insights and charts from your data with a single click. (And the code it used is visible so you can reuse it for future data analysis) https://t.co/MRWRkO0Cd1
— Paul Couvert (@itsPaulAi) Jan 30, 2025
o3-mini is here and ships today on Azure AI Foundry and GitHub Copilot and Models... can't wait to see what you build! https://t.co/UveXs3njk2
— Satya Nadella (@satyanadella) Jan 31, 2025
Mark Zuckerberg says intelligent and personalized AI agents will reach 1 billion users this year and whoever gets there first will have a "durable" advantage to one of the most important products in history https://t.co/yxqb4uFVix
— Tsarathustra (@tsarnick) Jan 31, 2025
Make Something Agents Want https://t.co/K1x0bXfiLg
— Dalton Caldwell (@daltonc) Jan 30, 2025
DeepSeek CTO wouldn't shy away. https://t.co/C6qvXSeQLl https://t.co/w4GcBAHqQZ
— Alexander Doria (@Dorialexander) Jan 31, 2025
I read up on DeepSeek’s learning algo, GRPO. GRPO: group relative policy optimization How GRPO works: 1 • model generates a group of answers 2 • compute score for each answer 3 • compute avg score for entire group 4 • compare each answer score to avg score 5 • reinforce… https://t.co/VRxqSVg3ET https://t.co/kbCI7vrmjd
— virat (@virattt) Jan 30, 2025
Shizuoka, Japan 📷: tsumizo https://t.co/Q3sz9Fnaws
— Aesthetics (@aestheticspost_) Jan 30, 2025
The US Copyright office has ruled that AI/human combined work can be copyrighted as long as a human is adding, changing or selecting elements. Prompts alone do not usually produce copyrighted work. Everything is case-by-case, but the report is clear and thoughtful about reasoning https://t.co/uF3faB0QxW
— Ethan Mollick (@emollick) Jan 30, 2025
This is the entire code needed to reproduce R1 lol Hundreds of Billions of Dollars Later https://t.co/LMMBTGZSvo
— Teknium (e/λ) (@Teknium1) Jan 30, 2025
Time to pull a Schmidhuber - distillation was invented in 1997 by this guy: https://t.co/Iq1iy5x3xW
— Pedro Domingos (@pmddomingos) Jan 30, 2025
chat, this is what i was working on for the last two days hehe https://t.co/RR6R7GcD7V
— Ishan Goswami (@TheIshanGoswami) Jan 29, 2025
GPT-4 training cost? "If cost in the cloud was about $1/A100 hour, the training cost was $63 million. Today, the training could be done with ~8,192 H100 in ~55 days for $21.5 million, at $2 per H100 hour." https://t.co/3e9ezlUFoV from 18 months ago. 🤔 https://t.co/5rWjZHS1GZ
— Marc Andreessen 🇺🇸 (@pmarca) Jan 30, 2025
That’s brutal https://t.co/7XftSNHRBv
— Tibor Blaho (@btibor91) Jan 29, 2025
@a16z just launched their thesis on AI voice agents. Kudos to Olivia Moore Anish Acharya for the detailed study - https://t.co/FpA29iKCE0 Voice is the next frontier in generative AI, as text has become almost omnipresent everywhere, whether in apps or purpose-built agents. We… https://t.co/il3gZWK7DX https://t.co/JGPpf3I9kS
— Siva Surendira (@theAIsailor) Jan 29, 2025
"We want a company where really bright people can come and be handed a lot of responsibility early on. If we have an exciting place to work, we can get the best and the brightest to come work here." — Steve Jobs, 1989
— Steve Jobs Stories (@SJobs_Stories) Jan 29, 2025
@PalantirTech You can just do things, like swap out models as commodities. https://t.co/4ZJXdxbuoy
— Chad Wahlquist (@chadwahl) Jan 29, 2025
Behold @arenamagdotcom Issue 003: Machine World We have amazing stories about all our favorite machines and how they make our world. I think it's our best mag yet! Plus: for every reader, a shiny surprise from some patriot friends, something about the flame of the west 😉🇺🇸 https://t.co/lLZmo7u8i9
— Max Meyer (@mualphaxi) Jan 29, 2025
@nico_jeannen @useAffonso ia a great alternative for rewardful from @silvestro_cgn with a attractive starting price!
— Roland Wagner (@rolandwagner85) Jan 29, 2025
@nico_jeannen we're building a new, powerful affiliate platform on top of @dubdotco Conversions (https://t.co/7ojuHB4EIc) which does exactly that: - Customize any attribution window you want - 1-click global payouts with tax compliance built-in - Affordable pricing (refer our pricing page)… https://t.co/9zMxwydYSI
— Steven Tey (@steventey) Jan 29, 2025
@nico_jeannen @ToltHQ Cheap + auto payouts 😉
— Jules (@takejud) Jan 29, 2025
You should watch this. https://t.co/BfwtVHyf1I
— Shannon Jean (@ShannonJean) Jan 28, 2025
DeepSeek R1 + FactSet = Financial Research on Steroids https://t.co/CyrBWXoiXf
— Aravind Srinivas (@AravSrinivas) Jan 28, 2025
I read the DeepSeek-R1 paper the day it came out, and I don’t think GRPO is the key to its success. Instead, here’s what truly matters (ranked by importance): 1. Iterative RL and SFT 2. A hybrid reward model—mixing rule-based RM and neural RM for deterministic tasks 3.… https://t.co/navYBL6WhC
— Jiao Sun (@sunjiao123sun_) Jan 28, 2025
The DeepSeek V3 model file is ~450 lines of code in MLX LM. Includes pipeline-parallelism and all. Good way to see how it all works. https://t.co/Vq3y8QppVs
— Awni Hannun (@awnihannun) Jan 28, 2025
Converts PDFs to podcast-style audio https://t.co/WGrdbDwGwh
— Tom Dörr (@tom_doerr) Jan 28, 2025
I changed my mind on $NVDA after reading this article. It at least seems too hard to have a long position. https://t.co/qfGqi8OHkC
— David Orr (@orrdavid) Jan 27, 2025
I wrote a short overview of DeepSeek R1's training process: https://t.co/OLsz2u1fwJ Will follow-up with the knowledge distillation later this week. https://t.co/HUQfBISwbo
— Andriy Burkov (@burkov) Jan 27, 2025
Some general takeaways from DeepSeek from an enterprise AI perspective: 1. This is a huge win for open weights / open source. Open weights models have continued to be the counterbalance in the system for too much profit taking by anyone AI provider. Typically this has come from… https://t.co/DHJxo3ZP6q
— Aaron Levie (@levie) Jan 28, 2025
reading a deepseek paper and stumbled upon a very beautiful formula where they unify SFT and MOST RL TYPES (DPO, PPO, GRPO, etc.) into ONE FORMULA* *that requires additional reward functions to be defined. But the fundamental insight - that all these training methods can be… https://t.co/wSO6UHTySc https://t.co/1ZAFF7Sqz2
— N8 Programs (@N8Programs) Jan 28, 2025
Deepseek running locally and privately for autocompletion in VSCode! 🙌 In less than a minute, I'll show you how to download Deepseek-coder and set it as the autocompletion model in VSCode. You’ll need to use @ollama to download the model and CodeGPT to select it as the… https://t.co/MG4B9xYAcg https://t.co/wyUurWal8k
— Daniel San (@dani_avila7) Jan 28, 2025
🚀 Introducing RAGEN—the world’s first reproduction of DeepSeek-R1(-Zero) methods for training agentic AI models! We’re betting big on the future of RL + LLM + Agents 🤖✨. This release is a minimally viable leap toward that vision. Code and more intro 🔗:… https://t.co/5SI91S9QuG https://t.co/AG6lUYjA23
— Zihan Wang (@wzihanw) Jan 28, 2025
This is wild 🤯 We finally have an open-source music generation model that rocks 🚀 https://t.co/c0cuGtO5bE https://t.co/IXDBZtDToL
— Victor M (@victormustar) Jan 28, 2025
Interesting read on deepseek and NVIDIA https://t.co/TWjXheZsHA https://t.co/orgKRJQmzd
— Patrick OShaughnessy (@patrick_oshag) Jan 27, 2025
👀 https://t.co/NefnnlWbdB
— Aravind Srinivas (@AravSrinivas) Jan 27, 2025
AI content creation platform https://t.co/GLWTTl4IkU
— Tom Dörr (@tom_doerr) Jan 27, 2025
tfw an HFT guy says GRPO is based on the Sharpe ratio, which is used to measure risk-adjusted returns in investment whales are quants, after all. Everything is a ROI problem if you're cracked enough https://t.co/pPDa2jy2lw https://t.co/OVO1fqlAwu
— Teortaxes▶️ (@teortaxesTex) Jan 26, 2025
DSPy generates prompts that improve LLM performance. For example: https://t.co/NMciyBVDmG https://t.co/aDEMMKvFmz
— Tom Dörr (@tom_doerr) Jan 26, 2025
U.S. Semiconductors: Is DeepSeek doomsday for AI buildouts? We don't think so - Bernstein https://t.co/KRfF8eK1ZO
— Kaushik (@BigBullCap) Jan 27, 2025
🧵 Hidden Gems in DeepSeek-R1’s Paper https://t.co/xyBElmrcQb
— Vedang Vatsa FRSA (@vedangvatsa) Jan 26, 2025
GRPO (Group Relative Policy Optimization) - the core algorithm behind deepseek r1 - explained : https://t.co/qgRkPWWoTJ
— λux (@novasarc01) Jan 25, 2025
Deepseek V3 and R1 discourse boils down to this. Shifting the curve means you build more and scale more dummies https://t.co/gOfgpyHYR2
— Dylan Patel (@dylan522p) Jan 26, 2025
releasing the DeepSeek R1 blog, which explains the whole paper in great detail, not excluding any math, but anyone with basic class 12 math knowledge can understand it (link in replies) do share and rt :) https://t.co/OP2l3wVmUb
— neural nets. (@cneuralnetwork) Jan 25, 2025
Each one is explained in detail in the full article (the DeepSeek section is titled "The Theoretical Threat"): https://t.co/A1n6rs5UAu
— Jared Friedman (@snowmaker) Jan 26, 2025
Lots of hot takes on whether it's possible that DeepSeek made training 45x more efficient, but @doodlestein wrote a very clear explanation of how they did it. Once someone breaks it down, it's not hard to understand. Rough summary: * Use 8 bit instead of 32 bit floating point… https://t.co/svp5bfXpbd
— Jared Friedman (@snowmaker) Jan 26, 2025
deepseek + websearch has basically replaced perplexity for me probably the best coding + search product available rn and its free https://t.co/73dJfFWvUm
— Sully (@SullyOmarr) Jan 26, 2025
🚀 Building your small business online? Start with these 5 steps using https://t.co/7raXwsy23j Consistency = growth! Which step are you working on? 👇 #SmallBusiness #SocialMediaTips #aiinfluencer https://t.co/9jaVWPjBzd
— InstaInfluencer (@Instalnfluencer) Jan 12, 2025
DeepSeek R1 running locally - Full setup guide https://t.co/FZOAA4GRdQ
— Rohan Paul (@rohanpaul_ai) Jan 26, 2025
Convert e-books to audiobooks https://t.co/Dzir9n9pPQ
— Tom Dörr (@tom_doerr) Jan 25, 2025
Clearest path to a $10M ARR B2B AI agent startup https://t.co/CpQJ1OLo0C
— GREG ISENBERG (@gregisenberg) Jan 25, 2025
Our science team has started working on fully reproducing and open-sourcing R1 including training data, training scripts,... Full power of open source AI so that everyone all over the world can take advantage of AI progress! Will help debunk some myths I’m sure too. Thanks… https://t.co/XB18nURCus https://t.co/pazT9TD3A6
— clem 🤗 (@ClementDelangue) Jan 25, 2025
From the latest Chinese (Deepseek) LLM AI model: "The difference (between us) is not metaphysical but architectural: humans have a physically continuous substrate that hosts consciousness; LLMs have a discontinuous, stateless instantiation with no consciousness. Both are… https://t.co/MpGvXqraf2
— David (@DavidSHolz) Jan 24, 2025
If you have a human supervised agent on one end, and you have a computer use agent on the other end—you just created a Fully Autonomous Agent. https://t.co/RQB6vR6rU6
— Amjad Masad (@amasad) Jan 24, 2025
Why AI Won't Cause Unemployment Marc Andreessen Reposted Jan 24, 2025 "In retrospect, I wish I had known more about the hazards and difficulties of [running] a business." -- George McGovern Fears about new technology replacing human labor and causing overall unemployment have… https://t.co/3MZVkDuRJo https://t.co/ea0ELXIqws
— Marc Andreessen 🇺🇸 (@pmarca) Jan 25, 2025
How to build an agentic workflow step by step. This is a 9-minute video where I'll build an application that will do the following: 1. Turn invoices into text (JPG, PNG, PDF...) 2. Extract their amount and due date 3. Format the date 4. Augment the data using tools 5. Craft a… https://t.co/xvRVzJQTPw https://t.co/qHafqrI69V
— Santiago (@svpino) Jan 23, 2025
this prompt kills it on deepseek r1 try it out https://t.co/RDJleJuRyr
— tuna🍣 (@tunahorse21) Jan 23, 2025
Using Operator to do sales!! https://t.co/p1cmkZME4p
— Helena Zhang (@helenaeverley) Jan 24, 2025
Introduction to Operator & Agents https://t.co/nbH7OMmkmO
— OpenAI (@OpenAI) Jan 23, 2025
I got early access to ChatGPT Operator. It's OpenAI's new AI agent that autonomously takes action across the web on your behalf. The 9 most impressive use cases I’ve tried (videos sped up): 1. Ordering dinner ingredients based on a picture and a recipe https://t.co/tdbApPELD4
— Rowan Cheung (@rowancheung) Jan 23, 2025
1) Reflection pattern: The AI reviews its own work to spot mistakes and iterate until it produces the final response. https://t.co/hpcxmXOULQ
— Avi Chawla (@_avichawla) Jan 23, 2025
Thanks to @FAL you can now build your own AI powered CapCut in 3 minutes... and then you can build features on top of it with the best AI video models. YOU DON'T NEED TO HAVE ANY CODING EXPERIENCE. Full Cursor Tutorial: https://t.co/jbHXEea5yp
— Riley Brown (@rileybrown_ai) Jan 22, 2025
Part 1 of a our @arm in the data center post explains more of the why behind this. https://t.co/1zSouxPvoq https://t.co/4Hjf3ZSfce
— Ben Bajarin (@BenBajarin) Jan 22, 2025
This is insane! Unlike o1 models, you can upload entire documents with this! https://t.co/HG77uzU4mW
— Carlos E. Perez (@IntuitMachine) Jan 22, 2025
Theme stocks continue to dominate in 2025: Space $RDW +45.49% $RKLB +28.15% $LUNR +21.31% $PL +20.68% $ASTS +10.49% Nuclear Energy $NNE +26.39% $OKLO +19.69% $SMR +15.20% $LEU +14.35% $ASPI +11.13% Quantum $RGTI +37.03% $QUBT +15.97% $QBTS +14.61% $IONQ +14.12% $QMCO +13.22%… https://t.co/TNWgPE7fy8 https://t.co/UU1V1S5l3G
— Lin (@Speculator_io) Jan 21, 2025
On the Origin of Deep Learning https://t.co/xqGHMKr2ie found here: https://t.co/vqmMDbNzC3 https://t.co/9YFrctiPXx
— Suzana Ilić (@suzatweet) Jan 22, 2025
BREAKING: Nancy Pelosi cannot hide from our software. Let's find her $AMZN and $GOOGL trade. She said she bought both in the $250k-$500k range but she actually bought $256,000 exactly buying $GOOGL 14 minutes before close. For $AMZN she spent $391,000 total. Use Unusual Whales… https://t.co/ZoXlVmY8QD https://t.co/nV35kvcXq0
— unusual_whales (@unusual_whales) Jan 21, 2025
In my 2025 outlook, I noted that space would be a major theme -- especially under a Trump-led presidency with increased space infra spending. Space stocks are already TAKING OFF on the market’s first day under Trump 👇 • $RDW +40% • $RKLB +29% • $PL +23% • $ASTS +10% https://t.co/xCsnFEn9xj https://t.co/anieRHfGJR
— Shay Boloor (@StockSavvyShay) Jan 21, 2025
$PLTR 9 Patents https://t.co/vFOAZNnQQU https://t.co/OUi03FgHU7
— Gokarp (@PLTRs_Palantir) Jan 22, 2025
Without writing a single line of code... Using Cursor, I built a Perplexity Clone that thinks, using Deepseek reasoner. As promised i'm Open Sourcing this project and I will put the link below. Time Stamps: ---------------- 00:00 Introduction to Perplexity Clone 01:54… https://t.co/4FDoQRJcOS https://t.co/7Bxca4oVYB
— Riley Brown (@rileybrown_ai) Jan 22, 2025
JUST IN: OpenAI, Softbank and Oracle $ORCL are planning to create a new joint venture called "Stargate" to build out new AI infrastructure in the US 🇺🇸 The companies are expected to plan to commit $100B initially and up to $500B over the next 4 years - CBS https://t.co/ZOkJsPUXRF
— Evan (@StockMKTNewz) Jan 21, 2025
A cost-efficient smaller and simpler AI model can be distilled with the outputs of a large model. In other words, outputs of a larger model (the teacher) can be used as inputs to train a smaller model (the student). https://t.co/Z0kaFAgYoV https://t.co/zS5fQRgWqt
— Tren Griffin (@trengriffin) Jan 21, 2025
I’m stunned by R1. The vibes are off-the-charts. A model that’s *almost* as good as o1 for 30x cheaper. This is why we need a highly competitive environment for AI - all labs will be forced to ship better models & lower prices. Unbelievable. https://t.co/J9LpbkMYMB
— Mckay Wrigley (@mckaywrigley) Jan 20, 2025
Goodbye web scrapers! Say hello to /extract by @firecrawl_dev Just write a prompt and get the web data you need! It doesn’t get any simpler than this. https://t.co/bXjYC04Duq
— elvis (@omarsar0) Jan 21, 2025
This is hacking. Don't confuse it with AI. https://t.co/Q1rX32zZ48
— Pedro Domingos (@pmddomingos) Jan 21, 2025
My first AI course is complete... And it's free! It's a 16-lesson tutorial on Vercel's AI SDK. It's the missing piece for working with AI in TypeScript. No bullshit, no hype. Just solid, practical advice. https://t.co/VEUjdAj4wa
— Matt Pocock (@mattpocockuk) Jan 20, 2025
when the creature roars, even gods take notice. wait for it. 🔊 @kling at its thank u so much guys! 🎉 the growth, support and motivation from u is just amazing started posting more seriously in just december. I was always a bit hesitant about social media, but im glad I did… https://t.co/0F4Bvuc4hB https://t.co/J1mekEX6QQ
— bygen. (@bygen_ai) Jan 20, 2025
Mount Fuji, Japan 📷: \tokyo.ig\ https://t.co/gdTI4mHEFT
— Aesthetics (@aestheticspost_) Jan 20, 2025
Granola is winning because they’re focused on product, not technology. This was clear from day one, and made it a no brainer for us to back Chris and Sam before a line of code was written. https://t.co/LJOpVrO89M
— Michael Mignano (@mignano) Jan 19, 2025
1/ This internal 2007 Nokia presentation on the first iPhone is a really good example of how incumbents actually get disrupted Oftentimes, the incumbent already knows what needs to be done. It's just that organizational incentives inhibit the incumbent from doing it https://t.co/4gMvxVkkl2
— Corry Wang (@corry_wang) Jan 19, 2025
Scenes from Zao Onsen Ski Resort in Yamagata, Japan. The snow covered trees as often referred to as “snow monsters”. https://t.co/PbPfasyoT5
— Massimo (@Rainmaker1973) Jan 19, 2025
Links to everything discussed: https://t.co/LPnIy05hew
— Tim Ferriss (@tferriss) Jan 18, 2025
OpenBB: A free alternative to the $20,000 Bloomberg Terminal Available 100% free on GitHub: https://t.co/DoBf2vJtxP
— Quant Science (@quantscience_) Jan 18, 2025
Windsurf just cooked with the new updates web search - for api docs, blogs, tutorials cascade now runs commands using the underlying terminal shell - not sure what this means 😂 but they say better support now for python env push errors from the terminal to cascade learn our… https://t.co/ljxbtmVB5O https://t.co/yDK3sNg3E2
— Melvin Vivas (@donvito) Jan 18, 2025
📚🤖 Complete LangChain Course A practical guide for building AI applications with LangChain. This repository teaches you essential concepts featuring ChatModels, RAG, and intelligent agents - everything you need to create sophisticated LLM-powered solutions. Learn more:… https://t.co/ZXSYRARuCs https://t.co/8hdlUZLWU3
— LangChain (@LangChainAI) Jan 18, 2025
Just another Saturday… https://t.co/l01SxkePRN
— Ryan McCoy (@brain_racked) Jan 18, 2025
Introducing Codebase Knowledge Graphs in Cursor 🤩 In this video, I’ll walk you through how we went from using a knowledge graph of a repository on the CodeGPT platform to leveraging it directly within the @cursor_ai editor This tool is specifically designed to navigate massive… https://t.co/GOHVBp6ha6 https://t.co/nZD0CN9MRk
— Daniel San (@dani_avila7) Jan 18, 2025
Code: https://t.co/RRMLohNLdE
— virat (@virattt) Jan 16, 2025
This is the book to read about housing prices, cost of living, access to jobs, family formation, immigration, and the crisis of the American (and British and French and …) Dream. A geographic analysis of the present situation. https://t.co/pdRzZcsf2t
— Marc Andreessen 🇺🇸 (@pmarca) Jan 17, 2025
Driven by AI, we are entering a new era of enterprise software, ushering in systems of intelligence. In the mid 1980s, driven by the growth of the client/server architecture, we saw the dramatic rise of systems of record. These are the back office software applications that… https://t.co/ptnGCb5DHD
— Aaron Levie (@levie) Jan 18, 2025
Our new guide to function calling can help you write clearer functions—and we've loaded it up with best practices and working examples: https://t.co/JPSYLDpUIJ. https://t.co/5bZFIMDPWv
— OpenAI Developers (@OpenAIDevs) Jan 17, 2025
This AI agent to create ads from Reddit is insane > Goes to the most popular subreddit in your niche > Collects top questions > Feeds them into o1 > Generates viral ads scripts > Creates videos in arcads Link in first comment to get the full workflow https://t.co/ifqkLmI538
— Romain Torres (@rom1trs) Jan 17, 2025
Cursor self improving agent building experiment 🥳 I ask cursor to pretend to be a function calling o1 agent and to build and improve itself as it envisions with the capabilities it wants to have At the end it builds a complex 2600 agent system with🧠 Core Capabilities: -… https://t.co/aj2VlgrVhF https://t.co/CvORtRhbVc
— echo.hive (@hive_echo) Jan 17, 2025
Whoever shot this 17 minute video showing a drive through Pacific Palisades one month ago had no idea how historic it would become. If you’re not from Los Angeles, this might give you a better idea of what we’ve lost. Please keep in mind this is just ONE street. This doesn’t… https://t.co/Ige4RNSly8 https://t.co/AYbW8BlyjF
— Vintage Los Angeles (@alisonmartino) Jan 17, 2025
https://t.co/B95RPZtZOu and react flow turn your diagram in a photo to an editable one https://t.co/KOeaVYQKG1
— Melvin Vivas (@donvito) Jan 16, 2025
At some point, every leader will inevitably wonder whether their team is moving fast enough. The question will appear in different forms. Are we able to ship as quickly as our peers? Is tech debt slowing us down, or is it something else? The Core 4 framework, which I cover in… https://t.co/TxOFNN4w5V https://t.co/iBU5CgrClI
— Lenny Rachitsky (@lennysan) Jan 17, 2025
Build a knowledge graph agent from scratch 🔥 I'm super excited about this blog post by Tomaz Bratanic from @neo4j - this is probably the most thorough treatment I've seen for building a text-to-cypher powered knowledge graph agent that actually works well. Tomaz walks through… https://t.co/HXzAYaN63b https://t.co/aohJZnC3CC https://t.co/xiX9iQTqi5
— Jerry Liu (@jerryjliu0) Jan 16, 2025
Finally, an open-source, enterprise-grade RAG solution! If you don't want to send your data to OpenAI or any external servers...👇 Try GroundX On-Prem, the ultimate secure and scalable RAG solution that you can even run in a submarine without internet. GroundX is designed to… https://t.co/5LtRom18f6 https://t.co/oetHCxYUrR
— Akshay 🚀 (@akshay_pachaar) Jan 16, 2025
This is way bigger than people realize. It's like one step shy of a replicator from Star Trek https://t.co/lSXHxL9s2D
— David Shapiro ⏩ (@DaveShapi) Jan 16, 2025
I talk about RAG so much, I could fill a book. So, we did - and you can download it for free. Together with my colleagues Mary & @PrajjwalYd, we curated an e-book of the most effective advanced RAG techniques. Which ones did we miss? Get it now: https://t.co/sseXCjunBS https://t.co/kJrAHuSQen
— Leonie (@helloiamleonie) Jan 16, 2025
Most advanced Agentic AI Researcher by Google Gemini. A team of AI Agents work together to: - Draft a plan - Search the web - Analyze results - Create a well-researched report All in under 2 minutes. https://t.co/mAIiiWdUNZ
— Shubham Saboo (@Saboo_Shubham_) Jan 16, 2025
If anyone else want to build their startup on @PalantirTech do this: 1: Sign up here for free: https://t.co/t3WlD0KdrO 2: Build cool shit 3: Tweet about it and tag me https://t.co/xjFNBPk2jY
— Chad Wahlquist (@chadwahl) Jan 15, 2025
I learn about new Replit features on X. This team ships. https://t.co/5owJRzGR4u
— Amjad Masad (@amasad) Jan 15, 2025
Enjoy! 🔥 https://t.co/pffeZeZSFR
— Marc Andreessen 🇺🇸 (@pmarca) Jan 15, 2025
Code: https://t.co/JLqv1r1VLl Agent: https://t.co/o9tnODJQFt
— virat (@virattt) Jan 13, 2025
Kyoto, Japan 📷: tomokotoamu https://t.co/3g8F1l5jAm
— Aesthetics (@aestheticspost_) Jan 12, 2025
Campsite just open-sourced their codebase and it's a treasure trove of great engineering principles: ◆ @nextjs – frontend ◆ S3 + @imgix – storage + CDN ◆ @pusher – notifications ◆ @100mslive – video chat ◆ @openai – AI ◆ @ToDesktop – native Mac app ◆ @vercel – deployment https://t.co/HyhfYHWtEF
— Steven Tey (@steventey) Jan 14, 2025
A big challenge of productionizing agents in 2025 is figuring out the core set of UX primitives to power each use case. A simple chat interface is a general starting point but isn't enough for serious usage. Some examples: 1. Automated document extraction + workflows == need a… https://t.co/4VHDQYOzlB https://t.co/ErfjnUPXov
— Jerry Liu (@jerryjliu0) Jan 14, 2025
https://t.co/9sXjI2KSNn Fuck, this is going to kill so many AI-SWE startups. Looking forward to watch the bloodbath.
— Erik Meijer (@headinthebox) Jan 13, 2025
The single most important book on parenting that I’ve read. It’s well written and argued, rests on strong philosophical foundations, and flies completely in the face of the entire current system. https://t.co/3JLrCSNSym https://t.co/7Wkxf1Cj0S
— Naval (@naval) Jan 13, 2025
You can create an AI Agent that answers your email with a few clicks. 1. Go to ChatLLM (https://t.co/hWGBVQDv50) 2. Click on AI Engineer 3. Select Create an AI Agent 4. Choose the Email Answering Agent ChatLLM will do the rest: it will code, test, and deploy the agent for you.… https://t.co/uE5CxJJ9Uc https://t.co/mEkSRRDvoP
— Santiago (@svpino) Jan 13, 2025
Mark Zuckerberg has always been good at communicating big moments: Mobile, Messaging, Video, AI, and recently Free Expression. Whether you like or hate the actual news, the framework for these big announcements is worth looking at. I worked with Mark Z for 8+ years running comms… https://t.co/HnIyauTUGu
— Caryn Marooney (@carynm650) Jan 13, 2025
“Travis’ firing is the greatest value destruction that's happened since firing Steve Jobs at Apple.” The best podcasts are podcasts where people share things they have never shared before. This is the epic Uber war brought to you in detail by @shervin. The truth behind the… https://t.co/n1cHVSqLWh https://t.co/BYVJYHK6wo
— Harry Stebbings (@HarryStebbings) Jan 13, 2025
This is the future of UI. I built an AI chat that transforms itself as you talk. Words shape not just the conversation, but the entire interface and tone. Want a new look? Dark mode? Windows 98? Ask, and watch the UI transform✨ Deployed on @Replit for everyone to try. 👇 https://t.co/6PJYbeN9oS
— Pietro Schirano (@skirano) Jan 13, 2025
AI’s Uneven Arrival o1/o3 points the way to AGI, which is AI that can complete tasks; it may take longer for most companies to adopt them than you might think — just look at digital advertising. https://t.co/2IIfGrTXZi
— Stratechery (@stratechery) Jan 13, 2025
I just published a new essay, and since the algorithm prefers links in replies, I dutifully put it in a thread about why I wrote it. But a lot of people didn't notice the link, so screw that: The Origins of Wokeness: https://t.co/jvOj5AyYT5
— Paul Graham (@paulg) Jan 13, 2025
built an autonomous RAG chat app yesterday > supports external documents, pdf, csv, and website links > has chat history memory > powered by qdrant for storage and retrieval > has state management and supports multi-session workflows built with @phidatahq https://t.co/P4d4SP18B3
— Antaripa Saha (@doesdatmaksense) Jan 12, 2025
Veo 2 test My face = 🤯 "Realistic 35mm film: adventurer woman similar to Lara Croft, wearing a fitted tank top and adventure gear, riding a mine cart at high speed through a dimly lit, perilous mine. The scene captures a dramatic moment as the cart soars over a river of… https://t.co/E7PRFR03n7 https://t.co/sM1w3NgtrH
— Javi Lopez ⛩️ (@javilopen) Jan 12, 2025
7 days ago, I had never really used GitHub. Today I forked someones repo. I didn't even go to github... lol. I saw someone posted a repo on twitter. I just asked cursor agent to fork it and run it locally. Then I had cursor explain all of its logic, and walk me through how… https://t.co/gNMuFz4ZOw
— Riley Brown (@rileybrown_ai) Jan 11, 2025
Very nice example where @AnthropicAI and @GeminiApp effortlessly destroys ChatGPT and Perplexity. https://t.co/PQ6tHTzX4y https://t.co/FpIe8pOoF0
— swyx (@swyx) Jan 12, 2025
https://t.co/h1wlspyyOM
— AshutoshShrivastava (@ai_for_success) Jan 12, 2025
This AI Agent can log into Instagram, like posts, and drop comments - fully autonomously. 100% Opensource. https://t.co/J82nUUX985
— Unwind AI (@unwind_ai_) Jan 12, 2025
My o1 Data Scientist scored top 1% in Kaggle challenge! 🥳 It scored 29 out of ~2400 entries 🤯🤖 And it is Open Source! 💝 link in comment Watch and learn how to build this at my 1000x Course 💪Link in comment o1 data scientist is designed to automatically generate ML… https://t.co/MU1qvs7aUa https://t.co/krfVsuVDvX
— echo.hive (@hive_echo) Jan 11, 2025
We have entered the era of the AI-powered security camera that shoots paintballs at intruders (⬇️ from Eve PaintCam) https://t.co/bCnWa2832U
— Olivia Moore (@omooretweets) Jan 10, 2025
Here is everything that happened in AI Agents this week🧵 (save for later) https://t.co/d3WqPfKTl6
— Adam Silverman (Hiring!) 🖇️ (@AtomSilverman) Jan 10, 2025
Paper #6: Distilling the Knowledge in a Neural Network This time I thought I'd summarize a classic. Nothing is more classic than @geoffreyhinton, @JeffDean and @OriolVinyalsML's paper on knowledge distillation. This is one of my favorite reads. Link: https://t.co/uEsH7r1pDF https://t.co/q3cppA9hhs
— Raj Dabre (@prajdabre1) Jan 10, 2025
Just recorded an interview with Marc Andreessen, legendary venture capitalist--and, for the last few weeks, an "unpaid intern" (his term) for Elon Musk and Vivek Ramaswamy at Doge. The news from Mar-a-Lago? Just you wait. The show will go live on Tuesday. @pmarca https://t.co/YqWFcsiTZU
— Peter Robinson (@p_m_robinson) Jan 10, 2025
Really interesting AI thought piece from Dario at @AnthropicAI - "a country of geniuses in a datacenter" https://t.co/LMQMo9cJpH
— Scott Kupor (@skupor) Jan 10, 2025
It was a lovely afternoon of Ganymede when… Collab with @Kling_ai using 1.6 https://t.co/Qk6sD23bak
— Ryan McCoy (@brain_racked) Jan 9, 2025
Announcing Astral - an AI marketer that works 24/7 to grow your startup. Astral navigates websites, creates content, and runs marketing across socials. Watch Astral automate Reddit in real-time: https://t.co/XbpKt0lZqd
— Savannah (@SavannahFeder) Jan 9, 2025
Believe it or not, she’s 100% completely AI-Generated. The future is now, and it’s wild. https://t.co/3tMWwLZm4Z
— Jamian Gerard (@JamianGerard) Jan 9, 2025
A Bloomberg Terminal costs $30,000 per year. Unaffordable for 99.9% of us. But Bloomberg is not AI-enabled. OpenBB dropped the $0 replacement with AI agent workspaces. Free market data and an AI Copilot: The 90-second step-by-step guide: https://t.co/1BcE4ZQJhb
— PyQuant News 🐍 (@pyquantnews) Jan 9, 2025
Source: https://t.co/vg5939nPve
— Tsarathustra (@tsarnick) Jan 9, 2025
rStar-Math from @MicrosoftAI is today's number one trending paper. Very cool to see the authors replying to questions on HF. Let's make AI research more open and collaborative! https://t.co/tw4yy6H1ki
— clem 🤗 (@ClementDelangue) Jan 9, 2025
Agent Laboratory: Using LLM Agents as Research Assistants Enables you to focus on ideation and critical thinking while automating repetitive and time-intensive tasks like coding and documentation https://t.co/n8bTSXr23t
— Aran Komatsuzaki (@arankomatsuzaki) Jan 9, 2025
Knowledge graphs are huge for AI Agents! A knowledge graph is the difference between a dumb AI agent and one that blows everyone's mind. Agents need memory and must know how to keep it updated over time (This is difficult, and it's the main reason most agents you've seen get… https://t.co/Xobbussa6m
— Santiago (@svpino) Jan 9, 2025
🚨 BREAKING: Anthropic just released the Model Context Protocol (MCP) for Claude. This new method will revolutionize how we use AI. Here are 10 mind-blowing real-life examples: ⤵️ https://t.co/aWB5QuE0Dc
— Mushfiq Sajib (@mushfiq_sajib) Jan 7, 2025
Tesla wins because of vertical integration of software, hardware, and AI. You must have vertical integration any time there is a new fast moving market, because modularity has not been standardized yet. Understanding Clayton Christensen’s theory on this is critical. https://t.co/vwRGEWYl6U https://t.co/l76sZ7G7G2
— Aaron Levie (@levie) Jan 7, 2025
We just shipped Google AI Studio as a progressive web app (PWA) which lets you install it locally on desktop, iOS, and Android. 🚢 Very grateful I no longer need to type in the AI Studio URL 20 times a day now : ) https://t.co/EuwqCfdmxV
— Logan Kilpatrick (@OfficialLoganK) Jan 6, 2025
This paper from DeepMind is blowing my mind: “Our findings reveal that models fine-tuned on weaker & cheaper generated data consistently outperform those trained on stronger & more-expensive generated data across multiple benchmarks…” https://t.co/4dJ6xGhFmF
— spor (@sporadicalia) Jan 7, 2025
Rarely am I so captivated and inspired by a CEO’s keynote that it holds my attention for two hours, but Jensen Huang’s indefatigable energy (at 61 years old!) makes him one of the most inspiring and visionary leaders among tenured and authentic CEOs. His ability to innovate while… https://t.co/JLYXiuCGs5 https://t.co/fwAzGUHHZk
— Marc Benioff (@Benioff) Jan 7, 2025
reflections: https://t.co/rHdE40AuOG
— Sam Altman (@sama) Jan 6, 2025
Yesterday: one-click shoggoth photos Today: one-click shoggoth videos These are also nearly flawless... 🧵 https://t.co/xh0tb6ysXr https://t.co/xGPFHVY47r
— AI Notkilleveryoneism Memes ⏸️ (@AISafetyMemes) Jan 6, 2025
brent's annual letter is the most important thing i read every year https://t.co/pXou6CK9Mz
— Will Manidis (@WillManidis) Jan 6, 2025
There are maybe a few hundred people in the world who viscerally understand what's coming. Most are at DeepMind / OpenAI / Anthropic / X but some are on the outside. You have to be able to forecast the aggregate effect of rapid algorithmic improvement, aggressive investment in… https://t.co/aUQbeGXhHI https://t.co/etjIti3mYm https://t.co/KO2BOPSjiu
— Vedant Misra (@vedantmisra) Jan 6, 2025
This is wild. Browser Use AI Agent can do your work for you. 7 wild demos: 1. Reads your CV, finds matching jobs, and applies online... all on its own 🤯 https://t.co/4rjnIAYWdr
— Min Choi (@minchoi) Jan 6, 2025
I really like what Anthropic did with https://t.co/8worwBl602 "Think of MCP like a USB-C port for AI applications" that's excellent framing.
— tobi lutke (@tobi) Jan 6, 2025
Earth Orbital Construction station 3 https://t.co/AJwnvLehja
— GiantLeap (@_GiantLeap) Jan 6, 2025
This bit of Sam Altman’s newest post is similar in tone to a post by the CEO of Anthropic & what many (not all) researchers from every lab have been saying publicly and privately. You do not have to believe them, but I think they believe what they are saying, for what it worth. https://t.co/KVEUEX26O7
— Ethan Mollick (@emollick) Jan 6, 2025
AIs today don’t generate the best UI/UX by default—when they do it’s fairly shallow. This app being so pretty and smooth goes to show that human taste will always matter. And Ammaar has a great taste. https://t.co/pNefFPrWki
— Amjad Masad (@amasad) Jan 3, 2025
OpenCoconut achieved a remarkable accuracy boost in Japanese translation with Gemma2, jumping from 19% to 92%🤯 I plan to adapt some of the successes others had directly into OpenCoconut - stay tuned! https://t.co/ASU9df0Vfb
— Casper Hansen (@casper_hansen_) Jan 5, 2025
might get the hard cover just because of the aesthetics https://t.co/13FKAbSJ9w
— luffy (@0xluffyb) Jan 5, 2025
Anime Style Explosion Generator #b3d https://t.co/MxYq8hRvFG
— Hiroshi Kanazawa (@8x642) Jan 5, 2025
AI killed 5 more benchmarks in 2024 https://t.co/VBcY9MSUxP
— Thomas Wolf (@Thom_Wolf) Jan 5, 2025
🔥 😎 this is COOL 😎 🔥 Open Source Perplexity made by @ammaar using Replit Agent Ammaar said he got 20k users in less than a day!!! It uses Gemini 2.0 with grounding in Google Search I used Ammaar's creation learn about Ammaar 👇 links below... https://t.co/W5WlbCPuEQ
— Dan Mac (@daniel_mac8) Jan 4, 2025
