Loading video...

Video Failed to Load

Go Home

MAI Models are live on AI/ML API! We used MAI-Thinking-1 to recreate a working version of the Windows XP interface. What looked like a clean one-shot was actually powered by a CrewAI agent crew under the hood — a Planner agent broke the UI into components, a Coder agent...

aimlapi's profile picture

AI/ML API

2,690 subscribers

539,624 views • 3 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Today, we announced the public preview of MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 on Microsoft Foundry, bringing our first-party AI models directly into the hands of developers. Read more:
0:26

Today, we announced the public preview of MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 on Microsoft Foundry, bringing our first-party AI models directly into the hands of developers. Read more:

Microsoft Azure

22,633 views • 5 months ago

BREAKING: Microsoft just launched its own AI model family and the coding model inside GitHub Copilot. MAI-Code-1-Flash. Shipped to every Copilot tier. I just tested it to build an app. Check out the video 👇 85.8% benchmark score. 256K context. 60% token reduction on hard tasks. Here are the other models. → MAI-Code-1-Flash (coding, free) → MAI-Thinking-1 (reasoning, 35B params) → MAI-Image-2.5 (image generation) → MAI-Transcribe-1.5 (speech, 43 languages)
0:33

BREAKING: Microsoft just launched its own AI model family and the coding model inside GitHub Copilot. MAI-Code-1-Flash. Shipped to every Copilot tier. I just tested it to build an app. Check out the video 👇 85.8% benchmark score. 256K context. 60% token reduction on hard tasks. Here are the other models. → MAI-Code-1-Flash (coding, free) → MAI-Thinking-1 (reasoning, 35B params) → MAI-Image-2.5 (image generation) → MAI-Transcribe-1.5 (speech, 43 languages)

Priyanka Vergadia

37,018 views • 3 months ago

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.
0:40

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.

🚨 AI News | TestingCatalog

46,535 views • 3 months ago

1/4 We’re releasing MAI-UI—a family of foundation GUI agents. It natively integrates MCP tool use, agent user interaction, device–cloud collaboration, and online RL, establishing state-of-the-art results in general GUI grounding and mobile GUI navigation, surpassing Gemini-2.5-Pro, Seed1.8, and UI-Tars-2 on AndroidWorld. To meet real-world deployment constrains, MAI-UI includes a full-spectrum of sizes, including 2B, 8B, 32B and 235B-A22B variants. We are publicly releasing two models: MAI-UI-2B and MAI-UI-8B.
4:48

1/4 We’re releasing MAI-UI—a family of foundation GUI agents. It natively integrates MCP tool use, agent user interaction, device–cloud collaboration, and online RL, establishing state-of-the-art results in general GUI grounding and mobile GUI navigation, surpassing Gemini-2.5-Pro, Seed1.8, and UI-Tars-2 on AndroidWorld. To meet real-world deployment constrains, MAI-UI includes a full-spectrum of sizes, including 2B, 8B, 32B and 235B-A22B variants. We are publicly releasing two models: MAI-UI-2B and MAI-UI-8B.

Tongyi Lab

855,711 views • 9 months ago

🚨 do you understand what just happened to Microsoft.. Microsoft just dropped seven of its own MAI models, trained from scratch with zero distillation, and said its custom-tuned models already match GPT-tier quality at 10x lower cost. The partner that pays OpenAI's bills is now quietly building the thing that replaces it. - MAI-Thinking-1 hits human-preference parity with Sonnet 4.6 and 53% on SWE Bench Pro, right next to Opus 4.6 - MAI-Code-1-Flash delivers Haiku-class coding at just 5B params and is already shipping inside GitHub Copilot - A custom MAI model tuned for Excel matched GPT-5.4 while running up to 10x more efficiently - Their real weapon isn't the weights, it's that they're already inside everyone's Office, Teams, and VS Code The hard part was never training a good model. The hard part is being the room everyone already works in. Microsoft owns the room.
1:40

🚨 do you understand what just happened to Microsoft.. Microsoft just dropped seven of its own MAI models, trained from scratch with zero distillation, and said its custom-tuned models already match GPT-tier quality at 10x lower cost. The partner that pays OpenAI's bills is now quietly building the thing that replaces it. - MAI-Thinking-1 hits human-preference parity with Sonnet 4.6 and 53% on SWE Bench Pro, right next to Opus 4.6 - MAI-Code-1-Flash delivers Haiku-class coding at just 5B params and is already shipping inside GitHub Copilot - A custom MAI model tuned for Excel matched GPT-5.4 while running up to 10x more efficiently - Their real weapon isn't the weights, it's that they're already inside everyone's Office, Teams, and VS Code The hard part was never training a good model. The hard part is being the room everyone already works in. Microsoft owns the room.

BuBBliK

171,865 views • 3 months ago

Actress Cai Wenjing (Elvira Cai) mentioned Mai Mai on her recent interview! 🎤: And there’s also your big-eyed cute girl, you called Zhao Jinmai on Hi6 before, right? 👩: She’s also my baby girl, but I haven’t seen her in a long time, she’s been too busy. 🎤: Everyone's curious about how you two became friends 👩: We acted together, and it felt like we’d known each other forever. At that time, Mai Mai was still quite young and preparing for her college entrance exams, so I shared some advice with her about what to pay attention to. She would also tell me about her preparations like, “Sis, I’m preparing this dance,” and I’d say, “It’s beautiful, you did a great job.” We had really in-depth conversations, about the industry, about growing up. I even recommended a fitness coach to her and encouraged her to work out. Later on, we didn’t see each other for quite a long time. Then we met again in Changsha and had a meal together with Mai Mai and her mom, it was really nice. When she went out to travel, she even came back and we had another gathering, and she brought me a little gift. She’s truly so sweet. Mai Mai is a very, very good girl #ZhaoJinmai 🌾 #赵今麦
0:47

Actress Cai Wenjing (Elvira Cai) mentioned Mai Mai on her recent interview! 🎤: And there’s also your big-eyed cute girl, you called Zhao Jinmai on Hi6 before, right? 👩: She’s also my baby girl, but I haven’t seen her in a long time, she’s been too busy. 🎤: Everyone's curious about how you two became friends 👩: We acted together, and it felt like we’d known each other forever. At that time, Mai Mai was still quite young and preparing for her college entrance exams, so I shared some advice with her about what to pay attention to. She would also tell me about her preparations like, “Sis, I’m preparing this dance,” and I’d say, “It’s beautiful, you did a great job.” We had really in-depth conversations, about the industry, about growing up. I even recommended a fitness coach to her and encouraged her to work out. Later on, we didn’t see each other for quite a long time. Then we met again in Changsha and had a meal together with Mai Mai and her mom, it was really nice. When she went out to travel, she even came back and we had another gathering, and she brought me a little gift. She’s truly so sweet. Mai Mai is a very, very good girl #ZhaoJinmai 🌾 #赵今麦

殷果 ❄️🌾

31,773 views • 5 months ago

AI Digest No.64: Living, Loving & Losing Control in the AI Era Here are 5 key stories shaping the AI landscape. 🌅 Microsoft Launches MAI-Image-1 Microsoft unveiled its first in-house AI image generator, MAI-Image-1 — a move to rival DALL·E 3 and gain full control over its creative AI stack. ⚖️ ChatGPT Bows Out of Medicine & Law OpenAI will stop offering medical and legal advice after regulatory pressure, instead redirecting users to verified experts. A major step toward AI accountability. 🛍 Pinterest Introduces “Shop with Pinterest” Pinterest’s new AI assistant helps users visualize outfits, compare styles, and shop smarter — turning the app into a full-on fashion marketplace. 💬 Teens Turn to AI for Therapy & Romance A new UK study shows 1 in 4 teenage boys use AI companions for emotional support or relationships. Experts warn of dependency and distorted intimacy. 📉 The AI Industry Is Running on FOMO Analysts say the AI boom may be an economic bubble — driven by hype, not sustainable adoption. “The most expensive fear of missing out in history.” Watch our new AI video edition and read the full digest on Medium
0:31

AI Digest No.64: Living, Loving & Losing Control in the AI Era Here are 5 key stories shaping the AI landscape. 🌅 Microsoft Launches MAI-Image-1 Microsoft unveiled its first in-house AI image generator, MAI-Image-1 — a move to rival DALL·E 3 and gain full control over its creative AI stack. ⚖️ ChatGPT Bows Out of Medicine & Law OpenAI will stop offering medical and legal advice after regulatory pressure, instead redirecting users to verified experts. A major step toward AI accountability. 🛍 Pinterest Introduces “Shop with Pinterest” Pinterest’s new AI assistant helps users visualize outfits, compare styles, and shop smarter — turning the app into a full-on fashion marketplace. 💬 Teens Turn to AI for Therapy & Romance A new UK study shows 1 in 4 teenage boys use AI companions for emotional support or relationships. Experts warn of dependency and distorted intimacy. 📉 The AI Industry Is Running on FOMO Analysts say the AI boom may be an economic bubble — driven by hype, not sustainable adoption. “The most expensive fear of missing out in history.” Watch our new AI video edition and read the full digest on Medium

GT Protocol

37,574 views • 10 months ago

🐯: We like going for drives together. We've gone to Chiang Mai before because the scenery is nice. We also like driving around together at night. 🐯: Actually, we've talked about driving all the way from Bangkok to Chiang Mai someday. 💬: If your schedules lined up, would you invite him? 🐯: Maybe. I think we'd enjoy that kind of trip together. // i hope they get to do it and enjoy their trip, a vlog is a must naaa~ 555 joking.. just enjoy between the 2 of you is enough for me~ PERTH DEWY VOLUMING TINT #barenblissxPerth
1:00

🐯: We like going for drives together. We've gone to Chiang Mai before because the scenery is nice. We also like driving around together at night. 🐯: Actually, we've talked about driving all the way from Bangkok to Chiang Mai someday. 💬: If your schedules lined up, would you invite him? 🐯: Maybe. I think we'd enjoy that kind of trip together. // i hope they get to do it and enjoy their trip, a vlog is a must naaa~ 555 joking.. just enjoy between the 2 of you is enough for me~ PERTH DEWY VOLUMING TINT #barenblissxPerth

chuchu🐼♡

18,250 views • 3 months ago

Satya Nadella was asked why Microsoft, with a $175 billion AI budget, still doesn't have a frontier model. Satya Nadella: "We are well on our way building our MAI models our goal is to basically hill climb from the bottom, by the way, not distilling anything." "My advice is more like use all but be independent of all. I would pull out a model and see whether I can retain the eval. If I can't, that means you really are dependent on something that may or may not be yours." "If you're a hyperscaler, you're not a supplier to two model companies, that's not a business we need more." "I worked on Windows interrupt with Unix first. In fact, it was counterintuitive, right ? We used to think, oh my god, this means we'll be less used, except we were more used we were able to penetrate the enterprise primarily because we did that interrupt work." If owning the infrastructure layer was the Microsoft plan and winning move, why is the same company now racing to build its own frontier model ?
3:41

Satya Nadella was asked why Microsoft, with a $175 billion AI budget, still doesn't have a frontier model. Satya Nadella: "We are well on our way building our MAI models our goal is to basically hill climb from the bottom, by the way, not distilling anything." "My advice is more like use all but be independent of all. I would pull out a model and see whether I can retain the eval. If I can't, that means you really are dependent on something that may or may not be yours." "If you're a hyperscaler, you're not a supplier to two model companies, that's not a business we need more." "I worked on Windows interrupt with Unix first. In fact, it was counterintuitive, right ? We used to think, oh my god, this means we'll be less used, except we were more used we were able to penetrate the enterprise primarily because we did that interrupt work." If owning the infrastructure layer was the Microsoft plan and winning move, why is the same company now racing to build its own frontier model ?

Vikram M

712,353 views • 9 days ago

If freen aunt hadn’t tricked her to go for the pageant I’m afraid we won’t have met actress freen😭😭 🗣️: You joined Miss Teen Thailand, right? 🌷: Yeah, I was just a kid back then. 🗣️: Who encouraged you to enter the pageant? 🌷: My aunt. She convinced me to give it a try. I don’t know what she saw in me, but at the time…honestly, I did t have money at that time the reason I wanted to go was because I wanted to go to Chiang Mai. 🗣️: Then why didn’t you just buy a ticket and go? 🌷: Instead, I entered the pageant! It was going to be my first time visiting Chiang Mai, so all I could think was, “I have to go to Chiang Mai.” I didn’t really care what the competition was I just wanted to go to Chiang Mai.but then I unexpectedly made it into the Top 15 of Miss Teen Thailand. I went in with no expectations at all, and before I knew it, things started taking off from there. #srchafreen‌
0:33

If freen aunt hadn’t tricked her to go for the pageant I’m afraid we won’t have met actress freen😭😭 🗣️: You joined Miss Teen Thailand, right? 🌷: Yeah, I was just a kid back then. 🗣️: Who encouraged you to enter the pageant? 🌷: My aunt. She convinced me to give it a try. I don’t know what she saw in me, but at the time…honestly, I did t have money at that time the reason I wanted to go was because I wanted to go to Chiang Mai. 🗣️: Then why didn’t you just buy a ticket and go? 🌷: Instead, I entered the pageant! It was going to be my first time visiting Chiang Mai, so all I could think was, “I have to go to Chiang Mai.” I didn’t really care what the competition was I just wanted to go to Chiang Mai.but then I unexpectedly made it into the Top 15 of Miss Teen Thailand. I went in with no expectations at all, and before I knew it, things started taking off from there. #srchafreen‌

‎ ✧

59,401 views • 2 months ago

Orynth Product Review EP12 Herald lets you deploy a real AI agent in one click. No code, no API keys, no cost. Pick a template, connect Telegram, and your agent is live in under 60 seconds, running 24/7 with memory and modular skills. Powered by Nous Research, one of the most respected names in open-source AI. Built by gigaevm ($/acc), who got tired of agent platforms charging $50/month and locking you into one model, so he made the free version. A real product listed on Orynth with its own token solana:BT91Uvx89v6XvJKMn6GiwvPAxN9Lo2MPTVWkuGg5vory In this twelfth episode, we break down what Herald is building, the builder behind it, and why a free, instant AI agent deployer is worth a look.
2:28

Orynth Product Review EP12 Herald lets you deploy a real AI agent in one click. No code, no API keys, no cost. Pick a template, connect Telegram, and your agent is live in under 60 seconds, running 24/7 with memory and modular skills. Powered by Nous Research, one of the most respected names in open-source AI. Built by gigaevm ($/acc), who got tired of agent platforms charging $50/month and locking you into one model, so he made the free version. A real product listed on Orynth with its own token solana:BT91Uvx89v6XvJKMn6GiwvPAxN9Lo2MPTVWkuGg5vory In this twelfth episode, we break down what Herald is building, the builder behind it, and why a free, instant AI agent deployer is worth a look.

Orynth News

16,871 views • 3 months ago

AI power users are here and the case for agent-friendly software is materializing. Aaron Levie on agents outnumbering people 1000x: "If you start to imagine that we all have to build software for agents… we spend as much time now thinking about the agent interface to our tool as we do the human interface." "If you have a hundred or a thousand times more agents than people, then your software has to be built for agents… it's going to be through an API or a CLI or MCP." "What if you give a coding agent access to your SaaS tools and a coding agent access to your knowledge work, workflows and context?" "It can actually code its way or use APIs through whatever task it's trying to achieve." "That appears to be like a paradigm that is starting to compound… and I actually think it kind of makes sense as the ultimate manifestation of this stuff." Aaron Levie
1:23

AI power users are here and the case for agent-friendly software is materializing. Aaron Levie on agents outnumbering people 1000x: "If you start to imagine that we all have to build software for agents… we spend as much time now thinking about the agent interface to our tool as we do the human interface." "If you have a hundred or a thousand times more agents than people, then your software has to be built for agents… it's going to be through an API or a CLI or MCP." "What if you give a coding agent access to your SaaS tools and a coding agent access to your knowledge work, workflows and context?" "It can actually code its way or use APIs through whatever task it's trying to achieve." "That appears to be like a paradigm that is starting to compound… and I actually think it kind of makes sense as the ultimate manifestation of this stuff." Aaron Levie

a16z

42,152 views • 1 month ago

Microsoft AI CEO Mustafa Suleyman explains the exit plan behind the biggest partnership in AI. It began with a Satya Nadella message surfaced in the trial: "I don't want to be Intel and have OpenAI be Microsoft." The escape hardware is already shipping. "Partnerships like that don't last forever. OpenAI wants to be a trillion-dollar public company... They want to have the freedom to operate and be able to buy compute from all sorts of other places, build their own compute, partner with whoever they want." "We've made the decision the early part of last year, which obviously informed all the contract negotiations, which then all got resolved and signed in October." "We now are able to manufacture and ship a chip that is 30% cheaper than a GB200 inside of our own clusters." "The MAI-Thinking-1 model that we've just released actually delivers 1.4x performance per watt improvement on top of the 30% improvement that you get from running on a Maia 200, once we co-optimize the models for our tasks." Microsoft watched Windows commoditize Intel and is refusing the Intel seat this time: own silicon, own frontier model, co-designed end to end. The open question is which gap closes first: OpenAI building Microsoft's enterprise distribution, or Microsoft catching OpenAI's frontier models. - Mustafa Suleyman (Mustafa Suleyman), CEO of Microsoft AI, on Decoder with Nilay Patel (nilay patel)
4:18

Microsoft AI CEO Mustafa Suleyman explains the exit plan behind the biggest partnership in AI. It began with a Satya Nadella message surfaced in the trial: "I don't want to be Intel and have OpenAI be Microsoft." The escape hardware is already shipping. "Partnerships like that don't last forever. OpenAI wants to be a trillion-dollar public company... They want to have the freedom to operate and be able to buy compute from all sorts of other places, build their own compute, partner with whoever they want." "We've made the decision the early part of last year, which obviously informed all the contract negotiations, which then all got resolved and signed in October." "We now are able to manufacture and ship a chip that is 30% cheaper than a GB200 inside of our own clusters." "The MAI-Thinking-1 model that we've just released actually delivers 1.4x performance per watt improvement on top of the 30% improvement that you get from running on a Maia 200, once we co-optimize the models for our tasks." Microsoft watched Windows commoditize Intel and is refusing the Intel seat this time: own silicon, own frontier model, co-designed end to end. The open question is which gap closes first: OpenAI building Microsoft's enterprise distribution, or Microsoft catching OpenAI's frontier models. - Mustafa Suleyman (Mustafa Suleyman), CEO of Microsoft AI, on Decoder with Nilay Patel (nilay patel)

Karl Mehta

56,899 views • 2 months ago

ONE PROMPT AND ONE BUTTON NOW BUILDS A FULL SYSTEM, DATABASE, BACKEND, AND FRONTEND INCLUDED Most people still picture AI as something you ask questions to, then copy the answer into your own code by hand. A builder wired Claude into Obsidian through MCP instead, so a single written prompt becomes the actual spec for the system. Claude breaks that prompt into database schema, UI components, and API endpoints on its own, then writes, runs, and fixes the code until it actually works. MCP is what makes this possible, it lets the agent see the filesystem, installed packages, and running services before it starts building anything. The output isn't a code snippet sitting in a chat window, it's a live local app running on its own server, ready to open in a browser. What used to need a backend developer, a DevOps engineer, and two weeks now finishes in under fifteen minutes. See the full architecture and prompt structure below👇
0:21

ONE PROMPT AND ONE BUTTON NOW BUILDS A FULL SYSTEM, DATABASE, BACKEND, AND FRONTEND INCLUDED Most people still picture AI as something you ask questions to, then copy the answer into your own code by hand. A builder wired Claude into Obsidian through MCP instead, so a single written prompt becomes the actual spec for the system. Claude breaks that prompt into database schema, UI components, and API endpoints on its own, then writes, runs, and fixes the code until it actually works. MCP is what makes this possible, it lets the agent see the filesystem, installed packages, and running services before it starts building anything. The output isn't a code snippet sitting in a chat window, it's a live local app running on its own server, ready to open in a browser. What used to need a backend developer, a DevOps engineer, and two weeks now finishes in under fifteen minutes. See the full architecture and prompt structure below👇

wast3

20,348 views • 2 months ago

#Aylio #政瞳 🐱: Actually, we'll be coming back to Bangkok again very soon. 🐺: Then I'll take you back to your 711 again, okay? 🐱:And we'll properly explore the city this time, because I think Bangkok's weather is actually really pleasant right now. 🐺: We can also visit lots of other places. 🐱: Right. And there's another place I really want to go... where was it again? 🐺: Where? 🐱: Chiang Mai. 🐺: Mmm. I especially want to take him to Chiang Mai because it's one of my favorite cities. It has a very artistic atmosphere, and the weather there is actually nicer than Bangkok's since it's farther north. It's at a slightly higher elevation, and a lot of things are grown there chocolate, as well as various fruits. I really like that place, and 🐱: Are there lots of motorcycles there? 🐺: Yeah, lots of motorcycles, lots of mountain roads, and Lots of greenery... and lots of elephants (LMAOO) BRO GOT COMPLETELY SOAKED BY BATHING AN ELEPHANT AND HE WANTS MORE 😭😭😭
1:10

#Aylio #政瞳 🐱: Actually, we'll be coming back to Bangkok again very soon. 🐺: Then I'll take you back to your 711 again, okay? 🐱:And we'll properly explore the city this time, because I think Bangkok's weather is actually really pleasant right now. 🐺: We can also visit lots of other places. 🐱: Right. And there's another place I really want to go... where was it again? 🐺: Where? 🐱: Chiang Mai. 🐺: Mmm. I especially want to take him to Chiang Mai because it's one of my favorite cities. It has a very artistic atmosphere, and the weather there is actually nicer than Bangkok's since it's farther north. It's at a slightly higher elevation, and a lot of things are grown there chocolate, as well as various fruits. I really like that place, and 🐱: Are there lots of motorcycles there? 🐺: Yeah, lots of motorcycles, lots of mountain roads, and Lots of greenery... and lots of elephants (LMAOO) BRO GOT COMPLETELY SOAKED BY BATHING AN ELEPHANT AND HE WANTS MORE 😭😭😭

Pookie🍊

16,623 views • 1 month ago

🌷: have concert “together” next year mai? 🗣️: “together” means… 🌷: …more than one, which is two. Well, what should I reply? (question was asked from a Vietnamese fan). 🗣️: what should be the answer? 🌷: I’m thinking if I can do it. Let me think with “together” dai mai? thank you. 🗣️: I saw somebody ask you about a series with “together” (since you’ll have TWOS next year) because you don’t usually respond much. so, would you like to do it? 🌷: by “interesting,” I mean that now i have my own company, and after having done different kinds of work, I feel like being a producer could be fun too. If one day i’ll have the chance to make my own series or take on other projects, it’s interesting. so, I replied “Maybe”. But first, let things grow and see what opportunities come our way.
1:55

🌷: have concert “together” next year mai? 🗣️: “together” means… 🌷: …more than one, which is two. Well, what should I reply? (question was asked from a Vietnamese fan). 🗣️: what should be the answer? 🌷: I’m thinking if I can do it. Let me think with “together” dai mai? thank you. 🗣️: I saw somebody ask you about a series with “together” (since you’ll have TWOS next year) because you don’t usually respond much. so, would you like to do it? 🌷: by “interesting,” I mean that now i have my own company, and after having done different kinds of work, I feel like being a producer could be fun too. If one day i’ll have the chance to make my own series or take on other projects, it’s interesting. so, I replied “Maybe”. But first, let things grow and see what opportunities come our way.

Por

175,305 views • 16 days ago

Our team was recently pitched by a well-known AI SDR company. Two problems came up: 1. The company only offered AI SDRs for outbound, we needed a solution to assist us on inbound processes. 2. The software was too complex and overpriced relative to our expected usage. So, I took a couple of weekends to build an early version for what we wanted: - Simple UI that syncs with GMail - Custom AI labeling for emails - AI Agent that drafts replies via CrewAI - Custom knowledge base for the Agent v - Slack inquiry tool for the Agent so it can ask us for clarifications anytime it has knowledge gaps With the help of AI, it's never been easier or more fun to take your workflows and turn them into full web apps. If you're interested in a released version of this, link below.
2:31

Our team was recently pitched by a well-known AI SDR company. Two problems came up: 1. The company only offered AI SDRs for outbound, we needed a solution to assist us on inbound processes. 2. The software was too complex and overpriced relative to our expected usage. So, I took a couple of weekends to build an early version for what we wanted: - Simple UI that syncs with GMail - Custom AI labeling for emails - AI Agent that drafts replies via CrewAI - Custom knowledge base for the Agent v - Slack inquiry tool for the Agent so it can ask us for clarifications anytime it has knowledge gaps With the help of AI, it's never been easier or more fun to take your workflows and turn them into full web apps. If you're interested in a released version of this, link below.

Horacio Lopez

24,281 views • 1 year ago

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.
0:23

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.

Avi Chawla

460,985 views • 3 months ago

i genuinely don't understand why i'm the only one using this... Andrej Karpathy, engineer at Anthropic, invented the concept of the LLM Wiki inside Obsidian i used it to build my own local knowledge base for generating AI videos on autopilot, here's how it works: - i sent to Claude Code hundreds of movies, ads, anime references - it ingested all of it and created smart links between each so it's easy for any agent to read it - whenever i want to make a video using Seedance 2.5 (coming soon on Higgsfield), the agent picks up the right format, composition and lens my whole video production pipeline lives in an AI agent with access to Higgsfield's suite: CLI, skills and all models (for image, video, voice and music) this is the absolute best way to work with AI agents in 2026... i teach you how to build the same in the article below:
1:59

i genuinely don't understand why i'm the only one using this... Andrej Karpathy, engineer at Anthropic, invented the concept of the LLM Wiki inside Obsidian i used it to build my own local knowledge base for generating AI videos on autopilot, here's how it works: - i sent to Claude Code hundreds of movies, ads, anime references - it ingested all of it and created smart links between each so it's easy for any agent to read it - whenever i want to make a video using Seedance 2.5 (coming soon on Higgsfield), the agent picks up the right format, composition and lens my whole video production pipeline lives in an AI agent with access to Higgsfield's suite: CLI, skills and all models (for image, video, voice and music) this is the absolute best way to work with AI agents in 2026... i teach you how to build the same in the article below:

Machina

53,232 views • 1 month ago

The entire AI industry is racing to build the smartest model. Satya Nadella just admitted that is not where the money is. The model is not the product. The harness is. That is the exact line. And it changes what Microsoft is actually competing on. OpenAI, Anthropic, Google, xAI, Meta every frontier lab is pouring hundreds of billions into training compute, chasing the next capability jump. Each betting that raw model intelligence is the moat. Microsoft is doing the opposite. It is building the harness the orchestration layer that sits above the model, connecting it to tools, data, permissions, sub-agents, and enterprise workflows. And it is letting OpenAI, Anthropic, and MAI compete to plug into it. "You need the model. But the model is not the product. The harness is." So do the math on what a harness actually does. A raw model dropped into an enterprise answers questions. That is a chatbot. A harness turns that same model into an agent that reads the SharePoint, edits the ERP entry, pulls the GitHub PR, updates Salesforce, and files the Excel report with the right permissions, the right audit trail, and the right sub-agent for each sub-task. The model provides the intelligence. The harness converts intelligence into work. Now here's where it gets interesting. "Even the best model in the world will feel broken without a great harness. And an okay model with a great harness can feel like magic." If that is true, the enterprise buyer is not buying model quality. The enterprise buyer is buying the harness. Which means model quality becomes a commodity input over time, and harness quality becomes the sustainable moat. Compare that to the strategy the entire frontier lab industry is executing. Everyone else is chasing the numerator raw intelligence. Almost nobody at scale is racing to build the denominator the orchestration layer that determines whether that intelligence can actually be deployed profitably inside a real company. The frontier model race has a 10 to 20 percent chance of producing a single dominant winner. Nadella just told the industry he does not need to be that winner. If OpenAI wins, Microsoft wins. If Anthropic wins, Microsoft wins. If MAI wins, Microsoft wins. If someone Microsoft has never heard of trains a better model in 2027, Microsoft still wins. Because the compute they train on, the harness they get plugged into, the enterprise contracts they get delivered through, and the products they sit inside are all Microsoft. He is not building the best AI model. He is building the layer that the best AI model has to run on to make anyone money. I wonder which position looks more valuable in ten years.
3:19

The entire AI industry is racing to build the smartest model. Satya Nadella just admitted that is not where the money is. The model is not the product. The harness is. That is the exact line. And it changes what Microsoft is actually competing on. OpenAI, Anthropic, Google, xAI, Meta every frontier lab is pouring hundreds of billions into training compute, chasing the next capability jump. Each betting that raw model intelligence is the moat. Microsoft is doing the opposite. It is building the harness the orchestration layer that sits above the model, connecting it to tools, data, permissions, sub-agents, and enterprise workflows. And it is letting OpenAI, Anthropic, and MAI compete to plug into it. "You need the model. But the model is not the product. The harness is." So do the math on what a harness actually does. A raw model dropped into an enterprise answers questions. That is a chatbot. A harness turns that same model into an agent that reads the SharePoint, edits the ERP entry, pulls the GitHub PR, updates Salesforce, and files the Excel report with the right permissions, the right audit trail, and the right sub-agent for each sub-task. The model provides the intelligence. The harness converts intelligence into work. Now here's where it gets interesting. "Even the best model in the world will feel broken without a great harness. And an okay model with a great harness can feel like magic." If that is true, the enterprise buyer is not buying model quality. The enterprise buyer is buying the harness. Which means model quality becomes a commodity input over time, and harness quality becomes the sustainable moat. Compare that to the strategy the entire frontier lab industry is executing. Everyone else is chasing the numerator raw intelligence. Almost nobody at scale is racing to build the denominator the orchestration layer that determines whether that intelligence can actually be deployed profitably inside a real company. The frontier model race has a 10 to 20 percent chance of producing a single dominant winner. Nadella just told the industry he does not need to be that winner. If OpenAI wins, Microsoft wins. If Anthropic wins, Microsoft wins. If MAI wins, Microsoft wins. If someone Microsoft has never heard of trains a better model in 2027, Microsoft still wins. Because the compute they train on, the harness they get plugged into, the enterprise contracts they get delivered through, and the products they sit inside are all Microsoft. He is not building the best AI model. He is building the layer that the best AI model has to run on to make anyone money. I wonder which position looks more valuable in ten years.

Vikram M

21,463 views • 2 months ago