Browser Use, the tool making it easier for AI ‘agents’ to navigate websites, raises $17M | TechCrunch https://techcrunch.com/2025/03/23/browser-use-the-tool-making-it-easier-for-ai-agents-to-navigate-websites-raises-17m/
“Clearly written instruction manuals in plain text are the future of APIs. I suspect this effect will be multiplied as AI models will start to use more accessible/less error prone methods of accomplishing their goals, something you can already see happening if you use AI a lot.” / X https://x.com/emollick/status/1902738826822308100
“people love MCP and we are excited to add support across our products. available today in the agents SDK and support for chatgpt desktop app + responses api coming soon!” / X https://x.com/sama/status/1904957253456941061
Chinese Startup Behind Manus AI Agent Seeks $500 Million Valuation — The Information https://www.theinformation.com/articles/chinese-startup-behind-manus-ai-agent-seeks-500-million-valuation
PwC launches AI Agent Operating System for enterprises: PwC https://www.pwc.com/us/en/about-us/newsroom/press-releases/pwc-launches-ai-agent-operating-system-enterprises.html
“BREAKING: Microsoft just dropped the Model Context Protocol (MCP) for Copilot Studio. No more manual updates, no more endless maintenance—connect once, and your agents automatically stay fresh with real-time updates. With MCP, building agents just got easier: ⚡ Instant” / X https://x.com/FlowAltDelete/status/1902449616890822975
“🚨 Gemini 2.5 Pro Exp dropped and it’s now #1 across SEAL leaderboards: 🥇 Humanity’s Last Exam 🥇 VISTA (multimodal) 🥇 (tie) Tool Use 🥇 (tie) MultiChallenge (multi-turn) 🥉 (tie) Enigma (puzzles) Congrats to @demishassabis @sundarpichai & team! 🔗 https://x.com/alexandr_wang/status/1904590438469951873
“Gemini 2.5 Pro Experimental on Livebench 🤯🥇 https://x.com/OfficialLoganK/status/1904925675892728179
“Introducing Gemini 2.5 Pro Experimental! 🎉 Our newest Gemini model has stellar performance across math and science benchmarks. It’s an incredible model for coding and complex reasoning, and it’s #1 on the @lmarena_ai leaderboard by a drastic 40 ELO margin. Only a handful of https://x.com/OriolVinyalsML/status/1904583691566727361
“Gemini 2.5 Pro is an awesome state-of-the-art model, no.1 on LMArena by a whopping +39 ELO points, with significant improvements across the board in multimodal reasoning, coding & STEM. You can try it out now in AI Studio https://x.com/demishassabis/status/1904587103805006218
“Today we are launching 2.5 Pro! I think it’s the best model in the world. State-of-the-art reasoning and great vibes (+39 ELO gap on lmsys!) 2.5 Pro improves in coding, stem, multimodal, instruction following, and lots more. Available in AI Studio & the Gemini App! https://x.com/jack_w_rae/status/1904583894458110218
“Introducing Gemini 2.5 Pro, the world’s most powerful model, with unified reasoning capabilities + all the things you love about Gemini (long context, tools, etc) Available as experimental and for free right now in Google AI Studio + API, with pricing coming very soon! https://x.com/OfficialLoganK/status/1904580368432586975
Gemini 2.5: Our newest Gemini model with thinking https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/
“Gemini 2.5 now does a much better job than most AIs writing plots & characters, thanks in part to its ability to “think” through details in advance. A fun test: “play a game of Fiasco (think a Coen Brothers movie, but as an RPG), make it interesting in terms of plot and prose” https://x.com/emollick/status/1904656593083396541
Google is rolling out Gemini’s real-time AI video features | The Verge https://www.theverge.com/news/634480/google-gemini-live-video-screen-sharing-astra-features-rolling-out
“Deep Dive Video: Complex image editing used to take hours — now Google’s Gemini 2.0 turns advanced ComfyUI & Photoshop workflows into simple text prompts. Here’s exactly how to try it (completely free). Chapters: 00:00 Conversational Editing with Google’s Multimodal AI 00:53 https://x.com/bilawalsidhu/status/1903172149764034605
“Google’s new Gemini 2.5 Pro Experimental takes the #1 position across a range of our evaluations that we have run independently Gemini 2.5 Pro is a reasoning model, it ‘thinks’ before answering questions. Google has released it as an experimental API in AI Studio only, and has https://x.com/ArtificialAnlys/status/1904923020604641471
“BREAKING: Gemini 2.5 Pro is now #1 on the Arena leaderboard – the largest score jump ever (+40 pts vs Grok-3/GPT-4.5)! 🏆 Tested under codename “nebula”🌌, Gemini 2.5 Pro ranked #1🥇 across ALL categories and UNIQUELY #1 in Math, Creative Writing, Instruction Following, Longer https://x.com/lmarena_ai/status/1904581128746656099
The “think” tool: Enabling Claude to stop and think \ Anthropic https://www.anthropic.com/engineering/claude-think-tool
OpenAI adopts rival Anthropic’s standard for connecting AI models to data | TechCrunch https://techcrunch.com/2025/03/26/openai-adopts-rival-anthropics-standard-for-connecting-ai-models-to-data/
“Great to see OpenAI adding support for MCP! It’s amazing to think that in less than 4 months, MCP has gone from just an idea we had at Anthropic on how to make integrations easier for devs to the industry standard for all AI app integrations.” / X https://x.com/alexalbert__/status/1904965223448006805
“We are introducing answer modes in Perplexity to make the core search product even better for verticals: travel, shopping, places, images, videos, jobs. The next step is to get super precise that you don’t have to press on these tabs. Available on web for now. Mobile soon. https://x.com/AravSrinivas/status/1904571071250260110
“We’re working on an updated version of Deep Research (shipping next week, hopefully). This will throw even more compute, think longer, present more detailed answers, use code execution, and render in-line charts. Example: Asking to reason how AWS can hit a $10T market cap. https://x.com/AravSrinivas/status/1902876897773760577
“17/ @n_sri_laasya just replaced her real estate broker with an AI agent for $0.01  – Finds properties  – Breaks down mortgage options  – Explains neighborhoods Built using @OpenAI’s Agents SDK + @AgentOpsAI for real-time observability https://x.com/AtomSilverman/status/1904972452045947262
Introducing Researcher and Analyst in Microsoft 365 Copilot | Microsoft 365 Blog https://www.microsoft.com/en-us/microsoft-365/blog/2025/03/25/introducing-researcher-and-analyst-in-microsoft-365-copilot/
“LlamaExtract is now in public beta 🔥- the leading, genAI-native agent for structured document extraction. We adapt the latest models and tune them so that you can structure even the most complex documents (financial reports, invoices, resumes, and more) in a highly accurate, https://x.com/jerryjliu0/status/1902880391578653176
“☎️Vodafone transforms data operations with AI using LangChain and LangGraph I’ve been super impressed working with the Vodafone team on their GenAI initiatives. Read here about how they stood up a multi-agent system performance metrics monitoring https://x.com/hwchase17/status/1904216034095333392
“New agents benchmark: CollaborativeAgentBench is the first benchmark studying collaborative LLM agents that work with humans across multi-turn collaboration on realistic tasks in backend programming & frontend design ⬇️ https://x.com/AIatMeta/status/1903146899458363442
AI Companions: Weird and Creepy or Worth Your Attention? – Genies https://genies.com/blog/ai-companions-weird-and-creepy-or-worth-your-attention
“As part of this work, we’re also releasing SWEET-RL, a novel RL algorithm for long-horizon & multi-turn tasks which can perform better credit assignments. Our experiments demonstrate that SWEET-RL achieves a 6% absolute improvement in success & win rates on” / X https://x.com/AIatMeta/status/1903146901068988473
“Last week we launched 6 integrations in 5 days. Here’s what you need to know! https://x.com/AgentQL/status/1901727648268558820
“Excited to announce AutoGen Studio v0.4.2 release updates! ✅ Improved Gallery with easier UI for editing tools, teams, agents & models ✅ Component validation & testing with real-time error highlighting ✅ Token streaming for better dev experience ✅ Side-by-side session https://x.com/pyautogen/status/1901712819793621027
“New Post: Exploring Model Context Protocol (MCP) With Spring AI https://x.com/baeldung/status/1900577669726511592
“New vibe coding course in collaboration with @Replit! Taught by Michele Catasta (@pirroh) and Matt Palmer (@mattppal), “Vibe Coding 101 with Replit” is available now. AI coding agents can generate code—but building real applications with them takes more than just prompting. It https://x.com/DeepLearningAI/status/1904918900380233806
“I found a prompt to make vibe coding way better. It’s OK if I code by vibes, but I don’t want my agents to be doing it (and they do it too often). https://x.com/gneubig/status/1904186575732253008
“CodeAct is a cool alternative to ReAct tool calling architectures: instead of using the built in tool calling parameters in the APIs, get the LLM to write code to call tools one big benefit – easily allows for describing a sequence of LLM calls” / X https://x.com/hwchase17/status/1904918196085547170
“Linear just launched expanded capabilities for bots: ✅ @mention bots ✅ assign tickets to bots Very low latency. Very smooth. Exactly what you’d expect from @linear 🔥 https://x.com/mathemagic1an/status/1904293320840655249
“✷ Collaborate with agents in @linear ✷ Excited to partner with @linear to launch our @codegen integration! ⚡️ > “Hey @codegen, can you solve this ticket? > Close any duplicates btw.” I haven’t started a task from scratch in days. Absolutely wild. https://x.com/mathemagic1an/status/1904293319297179871
“Computer use agents now available in LangGraph TypeScript, along with Python:” / X https://x.com/LangChainAI/status/1904932725989179675
“Debugging w/ LangGraph Studio LangGraph Studio is an IDE for visualizing + debugging agents. Here, we show two workflows to 1) isolate errors and 2) pause + review at specific steps. In both, we show how to make + test code changes with Studio. 📽️ https://x.com/LangChainAI/status/1904923672743469504
“Cursor now able to use 80% of Figma via MCP https://x.com/LinusEkenstam/status/1901381258694852973
MCP (Model Context Protocol): Simply explained in 5 minutes https://read.highgrowthengineer.com/p/mcps-simply-explained
“Model Context Protocol (MCP) in 60 seconds ⏰ Why I’m excited: 🚀 Thousands of MCP servers have been created over the past few months. 💡 MCP clients like IDEs, chat apps, and more are now integrating with these servers. https://x.com/gethackteam/status/1901330417011069090
“Model Context Protocol (MCP) explained! https://x.com/stevenheidel/status/1904966320770384170
“Tech Stack for Together Chat: – @togethercompute for all the LLMs – @firecrawl_dev for web scraping – @brave for the search API – @ClerkDev for authentication – @helicone_ai for observability – @nextjs for my framework – @neondatabase for my DB – @vercel for hosting https://x.com/togethercompute/status/1904204864885755905
Vellum AI https://www.vellum.ai/
Visual Generative AI API Suite https://bria.ai/api
Sakana AI on X: “Sakana AI super-powers AI reasoning using Japan’s own Sudoku Puzzles! Read more here → https://t.co/Sxqnpi0TuV At @NVIDIAGTC, Llion Jones @YesThisIsLion announced the release of our new reasoning benchmark based on the modern variant Sudoku to challenge the AI community. We https://t.co/8LnE9EpjYg” / X
https://x.com/SakanaAILabs/status/1902913196358611278
“For their IDE plug-in for AI coding and testing, @QodoAI Gen 1.0 uses LangGraph + @AnthropicAI’s MCP to power dynamic, multi-step agentic coding workflows. Check out how they built it: https://x.com/LangChainAI/status/1902044311858168112
“Build production-ready AI agents with Python, TypeScript, or Ruby. Mix Python for AI, TypeScript for type safety, and Ruby for APIs in the same AI agent. Deploy it in minutes with visual agent debugging and zero infrastructure setup. 100% Opensource. https://x.com/Saboo_Shubham_/status/1900373074249474242
“11/ Build a Travel Planner Agent using OpenAI Agent SDK Does Flight Recommendation, Hotel Recommendation & Budget Analysis https://x.com/AtomSilverman/status/1904972439874085048
“🔥 🤖 🕵️♂️ INTRODUCING: Meta Agent 💥 this is by far the most exciting thing I’ve created using AI!!! >>> made using Manus and @windsurf_ai 🔑 Meta Agent is an AI agent that uses the OpenAI Agents SDK to GENERATE NEW AGENTS you simply tell Meta Agent in natural language the https://x.com/daniel_mac8/status/1900735621036937709
“I’ve been pondering for a while whether the weave-agent planner should be more like ReActTree ( https://x.com/jd_pressman/status/1904139443189252252
“AI agents are automating all invoice processing workflows – and it starts with really really good structured extraction 🔥 LlamaExtract lets you transform even the most complex invoices into standardized schemas, including the overall vendor details + line items. We adapt the https://x.com/jerryjliu0/status/1904328371867361469
“Just published: how we’re using LangGraph by @LangChainAI to build agentic workflows in Qodo Gen 1.0. 🧵👇 https://x.com/QodoAI/status/1901635792394436925
“Our @crewAIInc integration in @weave_wb is officially live! ✌️ Say goodbye to multi-agent AI debugging h̶e̶a̶d̶a̶c̶h̶e̶s̶ processes. Now track every agent, task, LLM call, latency, and cost—all unified in one powerful interface. https://x.com/weights_biases/status/1904524590988013951
“Claude now has a good version of web search, which is really useful, but it also has the same issue as all non-agentic (that is, not Deep Research) AI uses of web search or RAG, which is the model only gets a small amount of context from search & then sometimes over-extrapolates. https://x.com/emollick/status/1902848359561408936
“Now you can integrate Recall into your agent using Model Context Protocol (MCP). With so many tools and frameworks like Claude, Cursor, Mastra and more supporting MCP, it just got so much easier to build agents with onchain storage.” / X https://x.com/recallnet/status/1902115446746144998
“🔥 🕵️♂️ 🧑💻 install Meta Agent as a Python package Meta Agent is now available to install with pip as a Python package via PyPi >>> pip install danmac-meta-agent run via command line with: danmac-meta-agent or import into your code 👁️ my vision is to maintain the Meta Agent https://x.com/daniel_mac8/status/1902549719529738709
“I have been learning about AI Agent for a while, and I want to share with you the AI toy I have been building in the past few weeks. Model Smart Contract Protocol (MSCP) It is inspired by Model Context Protocol (MCP). MSCP is a standard protocol that enables LLM applications https://x.com/Web3Rickey/status/1901939563104514535
Former Meta executives raise $15 million for AI assistant startup | Reuters https://www.reuters.com/technology/artificial-intelligence/former-meta-executives-raise-15-million-ai-assistant-startup-2025-03-27/
Outreach founder Manny Medina has a new startup that helps AI agents get paid | TechCrunch https://techcrunch.com/2025/03/25/outreach-founder-manny-medina-has-a-new-startup-that-helps-ai-agents-get-paid/
“Here’s something that has improved the output a lot with any coding agent like Claude Code, Cursor, etc while trying to solve a hard issue or working on a complex feature: https://x.com/skirano/status/1904294140021543168
“This paper introduces PLAN-AND-ACT, to improve agent planning using separate Planner and Executor models and synthetic data. 📌 Decoupling planning from execution reduces complexity. It allows specialized models. 📌 Synthetic data generation addresses the data scarcity for https://x.com/rohanpaul_ai/status/1904917007142699481
“Folks, this Thursday 4pm IST we’re creating an AI agent that can onboard new employees and help them apply for leaves in 60 minutes flat using @OpenAI’s brand new agent SDK. https://x.com/s0h4m/status/1904551309476594087
“smolagents has reached 15k GitHub stars : it’s one of the fastest-growing agentic frameworks ever 🚀🚀 Don’t hesitate to try it out, we working with @avillanovamoral to add nice stuff like sandboxed code execution via E2B or Docker! https://x.com/AymericRoucher/status/1904219464263946480
“13/ @MSITTechNews is building a multimodal multi-agent system using @Azure AI Agent Service and @OpenAI Agent SDK https://x.com/AtomSilverman/status/1904972443724448149
“We built an AI Real Estate Agent with Firecrawl’s Deep Research API! 🔥 Just enter your location and preferences to get real-time, up-to-date apartment listings instantly. Powered by Claude 3.7 Sonnet and @firecrawl_dev. https://x.com/ericciarla/status/1904203105727901860
“8/ @daniel_mac8 used gitingest to download all the docs from @OpenAIDevs Agents SDK Excellent for providing to your AI Coding Agent as context if building with the OpenAI Agents SDK https://x.com/AtomSilverman/status/1904972433712701691
“🤖 As AI agents become more autonomous, we’re racing toward a critical choice: How much control should we surrender? New op-ed in @techreview from @mmitchell_ai @evijitghosh @sashamtl @GiadaPistilli on why keeping humans in the loop isn’t just safer—it’s essential. https://x.com/fdaudens/status/1904190228686115056
“You should use pass^k for evaluating agents and not pass@k 🚩 pass@k: Measures the probability that at least one of k independent solution attempts will succeed. pass^k: Measures the probability that an agent would succeed on all k independent attempts. Imagine a customer https://x.com/_philschmid/status/1904147086011940942
“Langgraph computer use agent! Use OpenAI’s new computer use model with Langgraph to get a computer use agent in only a few lines of code” / X https://x.com/hwchase17/status/1904589229856080084
“🏥🔗 Medical Supply Chain AI LangChain-powered system revolutionizing medical supply chains in low-income countries through intelligent monitoring and fraud prevention. Discover the future of secure medical supplies 👉 https://x.com/LangChainAI/status/1904201544305725749
“announcing slack operator: a simple web agent powered by openai’s new computer-use model i’m dropping the source code below so you can use this for your next web agent project want to try a live version? leave a comment and i’ll add you to our slack! https://x.com/pk_iv/status/1901654960733077629
“🆕 talk + essay: Agent Engineering https://x.com/swyx/status/1904256213661192405
“If you’re a founder spending $$$ on outbound… you need to see this. I built an AI agent that does outbound better than your sales team for less than a dollar. 📡 Scans X, GitHub, and blogs for devs using Agents SDK 💬 Drops a helpful reply pointing them to @AgentOpsAI for https://x.com/n_sri_laasya/status/1904342784360013940
“15/ INTRODUCING: Meta Agent  Meta Agent is an AI agent that uses the OpenAI Agents SDK to GENERATE NEW AGENTS you simply tell Meta Agent in natural language the agent you want & and Meta Agent generates it for you https://x.com/AtomSilverman/status/1904972447809716357
Otter Meeting Agent: Your New Voice-Activated Teammate | Otter.ai https://otter.ai/blog/otter-meeting-agent-your-new-collaborative-teammate
“FlowiseAI now supports MCP Tools (Model Context Protocol). In this video we have a look at adding MCP servers to AI agents, including a custom MCP server to manage TO DO items. @FlowiseAI https://x.com/leonvz/status/1901187983295844368
“✨ Klarna’s AI Assistant, powered by LangGraph and LangSmith, handles customer support tasks for 85 million active users — reducing customer resolution time by 80% ✨ 🛍️ Klarna’s flagship AI Assistant is revolutionizing the shopping and payments experience. Built on LangGraph https://x.com/LangChainAI/status/1904219446874604018
“🚨 We’re giving access to a bunch of powerful agents built using @OpenAI’s Agents SDK 🏡 Real estate assistant 📊 Finance tools 🧠 Wellness agents + more Come hang at Office Hours today at 4 PM PT on our Discord. 👉 Register here: https://x.com/AgentOpsAI/status/1903138138270372082
“The Future of AI Agents at Nvidia” / X https://x.com/AtomSilverman/status/1902472452716052633
“🔌 Today we’re launching MCP support in @Letta_AI , enabling seamless integration with @Anthropic’s Model Context Protocol for connecting AI agents to powerful external tools. https://x.com/Letta_AI/status/1900245055874773222
“🔌 MCP Adapters Released! Introducing our new TypeScript library that connects Anthropic’s MCP tools with LangChain.js & LangGraph.js, featuring multi-server support and seamless agent integration. 📦 Install now: npm i @langchain/mcp-adapters https://x.com/LangChainAI/status/1903159677845745736
“OpenAI just released Voice AI Agents with GPT-4o Audio model. You can now build amazing voice AI Agents using OpenAI Agents SDK in a few lines of Python code. https://x.com/Saboo_Shubham_/status/1902769759273250837
Apple’s AI strategy plagued by delays, Siri upgrade remains in limbo | TechSpot https://www.techspot.com/news/107160-apple-exec-candidly-discusses-siri-struggles-including-ai.html
“Introducing A1Mail – Email for AI Agents! 📬🤖 TLDR: With A1Mail you can create an email address then send and receive mail from that address for your AI Agent – without paying $12 per month per Gmail or wrestling with integration issues https://x.com/Pashpops/status/1902814965595246855
“Want to use OpenAI computer use model in a langgraph agent? This is the easiest way to do so!” / X https://x.com/LangChainAI/status/1904589968116420924
“I just replaced my real estate broker with an AI agent. 🏡 Finds properties 💸 Breaks down mortgage options 📍 Explains neighborhoods Found a cute apartment in SF under $0.01. Built using @OpenAI’s Agents SDK + @AgentOpsAI for real-time visibility into every step your agents https://x.com/n_sri_laasya/status/1903131249767772252
“Our ongoing and deep partnership with @NVIDIA spans the early days of Android to today’s cutting-edge AI collaborations across Alphabet. I’m really excited about the next phase of our partnership as we work together on agentic AI, robotics, and bringing the benefits of AI to more” / X https://x.com/sundarpichai/status/1902110575871848838
“A couple weeks ago, I jumped into all things MCP (Model Context Protocol) and built an MCP Server that tells an agent in @cursor_ai exactly how to deploy a website to @digitalocean without requiring any input or assistance from me! Here’s a short video demo of it in action! https://x.com/WadeWegner/status/1902390311093235943
Zapier MCP—Connect your AI to any app instantly https://zapier.com/mcp
“🚀Developers, imagine supercharging your content management and chat services with a @Box + @weaviate_io integration powered by Weaviate’s new Query Agent service! Here’s why you’ll love building with this recipe: 🔎Smarter Search: Weaviate’s vector magic turns your Box content https://x.com/BoxPlatform/status/1902408624414023880
“🔭 End-to-End OpenTelemetry support in LangSmith LangSmith now offers full E2E OTel support for applications built with LangChain or LangGraph. This enables: 🔹Unified observability to see app performance across your stack in one cohesive view 🔹Distributed tracing to track https://x.com/LangChainAI/status/1905315130134401133
“🚀 “Creating a Polkadot AI Agent with the OpenAI SDK” Build an autonomous agent that analyzes OpenGov proposals, fetches on-chain data, and delivers structured reports using LLMs, MCP & Polkadot APIs. 🔗 https://x.com/Cyphertux/status/1903463449557942679
“6/ @Saboo_Shubham_ will be creating Voice AI Agent tutorials using OpenAI Agents SDK. He was able to create 50+ AI Agents and RAG tutorials, 100% free and open source. https://x.com/AtomSilverman/status/1904972429195366652
Amazon’s Alexa Fund is now backing AI startups | TechCrunch https://techcrunch.com/2025/03/26/amazon-alexa-fund-invests-into-four-new-startups-as-it-plans-to-invest-more-into-ai-solutions/
“// Building Effective Agents using Agents SDK // Had a lot of fun going through the agentic workflow examples using OpenAI Agents SDK. I shared a getting-started guide that I’m working on for an upcoming short course. If you missed it, find it below. (bookmark for later) https://x.com/omarsar0/status/1904649151004217654
“@GroqInc API docs are now completely LLM-friendly with /llms.txt and /llms-full.txt files. This is the future of technical docs – easily accessible knowledge for models, not just humans. Add these to your model’s context window and let your AI agents build fast with our API. https://x.com/ozenhati/status/1902544634825330724
“NVIDIA announced major agentic AI initiatives at GTC 2025, focusing on infrastructure, models, and development tools & is working with Google & Oracle on agents. I summarized everything you need to know. 🧵 https://x.com/AtomSilverman/status/1902087634492780764
“10/ @Cyphertux is creating a Polkadot AI Agent with the OpenAI SDK; an autonomous agent that analyzes OpenGov proposals, fetches on-chain data, and delivers structured reports using LLMs, MCP & Polkadot APIs. https://x.com/AtomSilverman/status/1904972437927993616
“12/ @s0h4m + team are creating an AI agent that can onboard new employees and help them apply for leaves in 60 minutes flat using @OpenAI’s brand new agent SDK. https://x.com/AtomSilverman/status/1904972441887363232
“14/ 🚨 We’re giving access to a bunch of powerful agents built using @OpenAI’s Agents SDK @AgentOpsAI 🏡 Real estate assistant 📊 Finance tools 🧠 Wellness agents + more https://x.com/AtomSilverman/status/1904972445863538915
“2/ @n_sri_laasya built an Agents SDK agent that automates outbound sales using @BrowserUse_DEV + @AgentOpsAI for real-time tracing. – Scans X, GitHub, and blogs for relevant posts – Reply’s with relevant responses – Costs less than $0.01 per lead +more https://x.com/AtomSilverman/status/1904972420563578923
“Building a multimodal, multi-agent system using Azure AI Agent Service and OpenAI Agent SDK https://x.com/MSITTechNews/status/1903771206358384735
“🕵️♂️ used gitingest to download all the docs from @OpenAIDevs Agents SDK excellent for providing to your AI Coding Agent as context if building with the OpenAI Agents SDK download from the GitHub link below ⬇️ https://x.com/daniel_mac8/status/1903928172892078238
“Checkout Together AI’s new cookbook on using LangGraph in agentic RAG systems!” / X https://x.com/LangChainAI/status/1904981007423406566
“Build a Travel Planner Agent using OpenAI Agent SDK Does Flight Recommendation, Hotel Recommendation & Budget Analysis. Manages: • Agents • Handoffs • Guardrails https://x.com/PratikKadam_/status/1903006466744705191
“We’re redefining what’s possible with AI. With the release of our latest model, Command A, optimized for real-world agentic and multilingual tasks, we’re demonstrating our commitment to bringing enterprises AI that goes beyond the ordinary, and offers security & efficiency. https://x.com/cohere/status/1905301761193377964
“## the deep research x ai builder thesis most “generative ui” is bullshit – just toolcalling premade ui but i think that is about to change -now- CoT: 2 agent form factors have hit PMF recently: the prompt-to-app ai builder and the deep research agent i see an inevitable” / X https://x.com/swyx/status/1903121115310067794
“🔄 To tackle large-scale code migrations, Uber’s Developer Platform team uses LangGraph to build a network of agents and automate unit test generation 🔄 Uber has established a dedicated Developer Platform AI team to more deeply integrate agentic systems into their engineering https://x.com/LangChainAI/status/1904967944410661070
“We’ve integrated Claude’s Model Context Protocol (MCP) so that users and AI agents can now easily interact with Story. Register IP, purchase licenses, automate revenue distribution, and more with the help of an AI agent. A major step towards AI-native programmable IP. https://x.com/StoryProtocol/status/1901634176606216533
“AgentOps 0.4.5 just released includes extended support for OpenAI Agents framework, LangGraph callbacks and numerous QoL fixes. Thanks to the whole team for contributing to this one! https://x.com/ssslomp/status/1904324241476194429
“7/ @omarsar0 shared – Building Effective Agents using Agents SDK He went through the agentic workflow examples using OpenAI Agents SDK and shared a getting-started guide https://x.com/AtomSilverman/status/1904972431502299281
“Introducing Diamond, our agentic AI-powered code review companion. Diamond gives you instant, codebase-aware feedback that eliminates review cycles and helps you ship higher-quality code faster. Works with or without Graphite, free up to 100 PRs/month. https://x.com/withgraphite/status/1902012302188310741
“Since o1, and especially since DeepSeek-R1, improved “reasoning” has basically become the standard for new LLMs. Speaking of which, Gemini 2.5 Pro just came out as the latest reasoning model offering, which ends up at the top of most benchmarks (notably Humanity’s Last Exam). https://x.com/rasbt/status/1904940955192418555
“2.5 Pro will come to Advanced users in the @GeminiApp. 💬 Simply select it in the model dropdown on desktop and mobile. It will also soon be available on @GoogleCloud’s #VertexAI platform. Developers and enterprises can start experimenting now in @Google AI Studio → https://x.com/GoogleDeepMind/status/1904581166755123463
“Gemini models, both 2.5 Pro and 2.0 Flash, have the fastest output speed compared to leading models https://x.com/ArtificialAnlys/status/1904923026820653435
“Gemini 2.5 Pro is live in the WebDev Arena! You can even see Gemini’s progress against its past. ⚔️ ✏️ “Animate layered triangles or polygons rotating in opposite directions with changing hues, creating a mesmerizing kaleidoscope effect”. https://x.com/lmarena_ai/status/1905002138172112971
“Gemini 2.5 Pro is a very good model, seems like a real step forward, in both metrics and practical use. I think because it is labelled 2.5 and was sort of quietly rolled out, people may miss how big a jump it is, but discussions are making me think others are feeling similarly.” / X https://x.com/emollick/status/1904756949284983200
“Introducing Gemini 2.5 Pro Experimental. The 2.5 series marks a significant evolution: Gemini models are now fundamentally thinking models. This means the model reasons before responding, to maximize accuracy — and it’s our best Gemini model yet. Blog -” / X https://x.com/NoamShazeer/status/1904581813215125787
“Built on our state-of-the-art Gemma models, TxGemma can understand and predict the properties of small molecules, chemicals, proteins and more. This could help scientists identify promising targets faster, predict clinical trial outcomes, and reduce overall costs. https://x.com/GoogleDeepMind/status/1905274926665208043
“Think you know Gemini? 🤔 Think again. Meet Gemini 2.5: our most intelligent model 💡 The first release is Pro Experimental, which is state-of-the-art across many benchmarks – meaning it can handle complex problems and give more accurate responses. Try it now → https://x.com/GoogleDeepMind/status/1904579660740256022
“Gemini 2.5 is a very impressive model so far. For example, it is only the third LLM (after Grok 3 and o3-mini-high) to be able to pull off “create a visually interesting shader that can run in twigl-dot-app make it like the ocean in a storm,” and I think the best so far. https://x.com/emollick/status/1904700257822540076
“Gemini 2.5 Pro is the most skilled model in the world, please let us know if you have any issues using it! ♊️♊️♊️” / X https://x.com/zacharynado/status/1904641052096754156
“Gemini 2.5 Pro Released and it’s beating every other model across the board🤯 💡Includes reasoning with Chain of Thought 💻Top 1 on Aider code editing leaderboard 🆓The model is free for everyone to use https://x.com/casper_hansen_/status/1904590489128440163
“Today we’re releasing an experimental version of Gemini 2.5 Pro. 💡2.5 Pro shows strong reasoning and improved code capabilities, with state-of-the-art performance across a range of benchmarks. 📈It’s topped @lmarena_ai’s leaderboard by a huge margin https://x.com/Google/status/1904581629017735261
“The new Agents SDK from @OpenAI has turned heads in the industry.👀 In this video, I explore it, going into detail on how to build systems, configure agents, use guardrails, and tracing.🤯 I even demo how to integrate Gemini from @GoogleDeepMind.🧠 https://x.com/stefanjblos/status/1902768994152529965
“Google is winning by so much. We have always been the best in AI. 🔥 Gemini 2.5 pro is the best model in the world 👇” / X https://x.com/YiTayML/status/1904598794278494272
“Don’t forget I made a specific MCP for this, which you can download from our official Anthropic MCP server repo. When I first built this tool, I saw the performance improvement firsthand, so it’s awesome to see research confirming my early thoughts. https://x.com/skirano/status/1903152968288932085
“🚧Building MCP Connect: An open-source client for Model Context Protocol Servers 🔄Current features: – OpenAI models integration – Dynamic tool/resource discover – Async stdio transport – Multi-server connections – Server config compatible with claude, cursor. More features soon https://x.com/abiorhmangana/status/1902285378335232427
“Announcing Appwrite’s Model Context Protocol (MCP) server 🤖 With the MCP server, LLMs can directly interact with Appwrite’s API to perform actions. Which means, now you can use apps like Claude, Cursor or Windsurf editor to make updates to your Appwrite project. We’ve covered https://x.com/appwrite/status/1900176291552477478
“Google JUST dropped Gemini 2.5 Pro their “most intelligent AI model yet”. This new “thinking model” tops LMArena by a significant margin and crushes DeepSeek-R1, Grok 3 and Claude 3.7 in math, science and coding benchmarks. Ships with 1M token context (2M coming), can process https://x.com/bilawalsidhu/status/1904579366232695128
“The future of AI research just leveled up. 🔥 The Perplexity API Model Context Protocol (MCP) is here. They built an MCP server for Sonar, giving AI assistants real-time web search capabilities. Now, Claude can tap into Perplexity-powered search, delivering instant, accurate https://x.com/aaliya_va/status/1900120241751154924
“Looks like @AnthropicAI are the first releasing combined reasoning + tool use. From the video it looks: 1. Claude starts to reason 2. Generates a function call 3. Executes the function call 4. Continues reasoning with the function output 5. Repeats reasoning + tool use until https://x.com/_philschmid/status/1903019035765170419
“Very excited to welcome the newest member of our team at @databricks – Claude! We’ve partnered with our friends at @AnthropicAI to make Claude available to Databricks customers on all clouds. Great data intelligence requires great models, and Claude is here to help!” / X https://x.com/jefrankle/status/1904916403481694640
“If you are using Claude for coding give Gemini 2.5 Pro a try!” / X https://x.com/_philschmid/status/1905272290931061111
“Vibecoding a version of missile command that includes relativity. (probably need the video full screen) Still a ways to go, but an interesting start given that I just asked Claude Code to create the game and gave it feedback, it did the ideas. It also decided to add lesson plans https://x.com/emollick/status/1904333416645955944
“Making Model Context Protocol server work with doge gov @DOGE and @AnthropicAI Claude Code https://x.com/true_jae_yang/status/1902088605750989020
“LlamaCloud can be used as an MCP server! This allows you to bring up-to-the-second data into your workflow as a tool used by any MCP client. In this video we demonstrate how to use an existing LlamaCloud index as a data source for an MCP server used by Claude Desktop, which can https://x.com/llama_index/status/1905332760614764810
“🚀 Announcing mcp-clojure-sdk: A Clojure SDK for Model Context Protocol Servers! I’m excited to open-source our Clojure SDK for building Model Context Protocol (MCP) servers! https://x.com/vedang/status/1900979926758814014
“We built an AI that simulates how well your post will do on social media. Based on real time daily trends from @firecrawl_dev web data and the X API. 100% open source and live now: https://x.com/ericciarla/status/1904563705326764405
“We raised $3.5m to build the world’s only AI employee that handles all the work of influencer marketing Try it today for free. Also, we’re hiring engineers in San Francisco. https://x.com/tdoggyholhol/status/1902364308899819793
“Kinda nuts how easy it’s about to be to add voice support on the server to Agents using @OpenAIDevs Agents SDK 8 lines? https://x.com/ParkerRex/status/1902777846617063774
“4/ @ParkerRex shares how easy it is to add voice support on the server to Agents using @OpenAIDevs Agents SDK in 8 lines 🤯 https://x.com/AtomSilverman/status/1904972424942395875
“1/ MCP support now available in the OpenAI agents SDK! Thanks for sharing @_rohanmehta https://x.com/AtomSilverman/status/1904972418369876267
“MCP 🤝 OpenAI Agents SDK You can now connect your Model Context Protocol servers to Agents: https://x.com/OpenAIDevs/status/1904957755829481737
Model context protocol (MCP) – OpenAI Agents SDK https://openai.github.io/openai-agents-python/mcp/
“HELM has a new leaderboard: HELM Capabilities v1.0! We curated 5 challenging datasets (MMLU-Pro, GPQA, IFEval, WildBench, Omni-MATH) and evaluated 22 top language models: https://x.com/percyliang/status/1902890719985160471
“💡 Good few people knew Apify, even more knew web scraping, but everybody knew MCP. Our Ondra Urban demonstrated all these things last week at AI Dev Tool Demo Night at AWS GenAI Loft in San Francisco – showing how to use Model Context Protocol (MCP) to tap into 4,000+ Apify https://x.com/apify/status/1902031271963873639
“MCP (Model Context Protocol) is very popular recently. We have studied the feasibility of combining it with BIOS several months ago when it was proposed. The idea of our meta-action sequence comes from it. https://x.com/biosphere3_ai/status/1901487238271349061
“I just vibe coded a whole iOS app in Swift (without having programmed in Swift before, though I learned some in the process) and now ~1 hour later it’s actually running on my physical phone. It was so ez… I had my hand held through the entire process. Very cool.” / X https://x.com/karpathy/status/1903671737780498883
