Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Don't do RAG - One counter intuitive thing I learnt past few weeks - Load whole knowledge base into prompt is actually fast, cheap & more accurate than normal RAG pipeline I was building an MCP for reading external doc, I was able to generate right code example by...

jasonzhou1993's profile picture

Jason Zhou

30,601 subscribers

82,818 Aufrufe • vor 1 Jahr •via X (Twitter)

BildungWissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

10 Kommentare

Profilbild von Mark Fer
Mark Fervor 1 Jahr

What if you want to dive into thousands of sales calls, marketing content, etc?

Profilbild von Breadcrumb
Breadcrumbvor 1 Jahr

Looking to automate reporting? Use AI agents to turn spreadsheets to reports in minutes without any coding.

Profilbild von Imiël Visser
Imiël Visservor 1 Jahr

Gemini has 1M context window vs Claude 200k... 50 pages is nothing, but what is the cost vs RAG for volume processing?

Profilbild von Brandon Culver
Brandon Culvervor 1 Jahr

No offense, but RAG isn't for 50 pages, it's for 500k pages

Profilbild von josh
joshvor 1 Jahr

Have you tried using Figma MCP? I’m curious if it’s possible to load prebuilt components that the AI can use during the build.

Profilbild von AJ
AJvor 1 Jahr

this is a good stuff. I am thinking of is cursor is able to read npm packages well or not because with not so known packages its' did a really bad job manybe we can connect something like this to a package wdyt?

Profilbild von Ajay Jagota 🔑🏃‍♂️
Ajay Jagota 🔑🏃‍♂️vor 1 Jahr

This is brilliant TY 🙏 Love how context drives the performance and you can track it in helicone 👏

Profilbild von Sam Sung
Sam Sungvor 1 Jahr

Do you think this is the way to go for giving “memory” to AI? I’m working on a chatbot that I want to give memory, but worried that as conversations get longer (and users have more conversations) that the cost of processing it all will be too much (currently used Claude 3.5 sonnet with caching, but still it gets expensive - like 50 cents for a single, long conversation). I’ve been wondering how I can figure out this memory issue - RAG? Create summaries with cheaper models?

Profilbild von Kris
Krisvor 1 Jahr

We are working on big 1GB+ pdf DB. I think OCR once. Summarize to 50 pages. Send that plus relevant files to Gemini to answer in super context? Wanna join our Discord?

Profilbild von Boring traveler
Boring travelervor 1 Jahr

How about in local?

Ähnliche Videos

Those MCP totally 10x my Cursor workflow… - It reads my browser console / network log - It use Replicate to generate UI assets - It reads my supabase & figma ... Here I shared some of my fav MCP & how to install them, As well as the EASIST way to create your own MCP (Thanks to Cloudflare 's MCP-worker) 0:00 Cursor MCP overview 2:35 How to install common MCP 5:49 Let cursor read console/network log 7:24 Build yoru own MCP If you want to learn more about advanced AI coding workflow, join my community where I post tips weekly:
13:16

Those MCP totally 10x my Cursor workflow… - It reads my browser console / network log - It use Replicate to generate UI assets - It reads my supabase & figma ... Here I shared some of my fav MCP & how to install them, As well as the EASIST way to create your own MCP (Thanks to Cloudflare 's MCP-worker) 0:00 Cursor MCP overview 2:35 How to install common MCP 5:49 Let cursor read console/network log 7:24 Build yoru own MCP If you want to learn more about advanced AI coding workflow, join my community where I post tips weekly:

Jason Zhou

287,888 Aufrufe • vor 1 Jahr

I've been using Playwright MCP in my coding workflow for 2 use cases: 1. Cursor Self-iterate UI 2. AI-driven test automation With the recent image support, it is getting even better. ------ Here is how I do it: 0:00 Playwright intro 2:43 My Setup 6:00 Use case 1: Iterate UI 7:20 Use case 2: Automated testing If you want to learn AI coding beyond basics, Join AI Builder Club where I post tips weekly:
10:30

I've been using Playwright MCP in my coding workflow for 2 use cases: 1. Cursor Self-iterate UI 2. AI-driven test automation With the recent image support, it is getting even better. ------ Here is how I do it: 0:00 Playwright intro 2:43 My Setup 6:00 Use case 1: Iterate UI 7:20 Use case 2: Automated testing If you want to learn AI coding beyond basics, Join AI Builder Club where I post tips weekly:

Jason Zhou

21,886 Aufrufe • vor 1 Jahr

Better Cursor alternative? I've been playing with the new AI IDE Windsurf launched by Codeium past few days, and the result was impressive, particularly: 1. Code base understanding: One thing Windsurf did really well is its understanding of the whole code base & dependencies; This led to a big difference in terms of final code quality 2. Context aware: It actually knows everything I did in the editor, e.g. it automatically knows code edits I made without me prompt it; 3. Reflection: I noticed Windsurf has been finetuned to reflect and try to iterate/improve code it generated autonomously; This was a magic moment for me .... I shared key highlights i had and showcase my process of building a PDF converter product using windsurf 0:00 Platform overview 6:21 How does context aware engine works 10:45 Case study: Build PDF converter Sass ---- If you have any further question or want to get deep dive into more AI coding, you can join my community where I post tips weekly:
26:13

Better Cursor alternative? I've been playing with the new AI IDE Windsurf launched by Codeium past few days, and the result was impressive, particularly: 1. Code base understanding: One thing Windsurf did really well is its understanding of the whole code base & dependencies; This led to a big difference in terms of final code quality 2. Context aware: It actually knows everything I did in the editor, e.g. it automatically knows code edits I made without me prompt it; 3. Reflection: I noticed Windsurf has been finetuned to reflect and try to iterate/improve code it generated autonomously; This was a magic moment for me .... I shared key highlights i had and showcase my process of building a PDF converter product using windsurf 0:00 Platform overview 6:21 How does context aware engine works 10:45 Case study: Build PDF converter Sass ---- If you have any further question or want to get deep dive into more AI coding, you can join my community where I post tips weekly:

Jason Zhou

64,713 Aufrufe • vor 1 Jahr

New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic Anthropic and taught by Elie Schoppik Elie Schoppik, its Head of Technical Education. Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on. MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely. In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers. Here’s what you’ll do: - Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources - Learn the core components of the client-server architecture of MCP and the underlying communication mechanism - Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application - Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector - Create an MCP client inside your chatbot to dynamically connect to your server - Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown - Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients - Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications - Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work. Please sign up here!
2:43

New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic Anthropic and taught by Elie Schoppik Elie Schoppik, its Head of Technical Education. Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on. MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely. In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers. Here’s what you’ll do: - Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources - Learn the core components of the client-server architecture of MCP and the underlying communication mechanism - Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application - Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector - Create an MCP client inside your chatbot to dynamically connect to your server - Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown - Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients - Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications - Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work. Please sign up here!

Andrew Ng

142,010 Aufrufe • vor 1 Jahr

Seedance 2.0 vs Gemini Omni, tested under the same conditions 👀 I was working on a new storyboard. After generating it with Seedance 2.0, I gave the exact same prompt + storyboard reference + character reference to Gemini Omni as well. Result: Gemini Omni surprised me with style quality and got closer than I expected in prompt adherence. But Seedance still feels ahead for storyboard execution, motion energy, camera language and environmental interaction. Gemini looks good. Seedance feels directed. You can see them in the video and the storyboard sheet is shown at the end. Which one do you think is more useful for this kind of work?
0:29

Seedance 2.0 vs Gemini Omni, tested under the same conditions 👀 I was working on a new storyboard. After generating it with Seedance 2.0, I gave the exact same prompt + storyboard reference + character reference to Gemini Omni as well. Result: Gemini Omni surprised me with style quality and got closer than I expected in prompt adherence. But Seedance still feels ahead for storyboard execution, motion energy, camera language and environmental interaction. Gemini looks good. Seedance feels directed. You can see them in the video and the storyboard sheet is shown at the end. Which one do you think is more useful for this kind of work?

Kōda

34,741 Aufrufe • vor 2 Monaten

MCP now supports authentication using OAuth2. This is huge! I want to share a complete example (code) so you can learn how to use it. Here is the thing: You don't want to build agentic applications that access resources without proper guardrails. With authentication, you can: • Call third-party APIs on the user's behalf • Perform actions that should be traceable • Enforce different permissions for different roles Here is an MCP-based web shop demo, built with WorkOS AuthKit and Next.js. With this example, you can do the following: • Connect to an MCP server • Authenticate using AuthKit • Order a T-shirt using MCP tools • View order details if you are an admin This GitHub repository contains everything you need: The WorkOS team has one of the best authentication platforms in the world. If you haven't looked into AuthKit, take a look at the code of this demo. Thanks to the WorkOS team for sponsoring this post.
5:10

MCP now supports authentication using OAuth2. This is huge! I want to share a complete example (code) so you can learn how to use it. Here is the thing: You don't want to build agentic applications that access resources without proper guardrails. With authentication, you can: • Call third-party APIs on the user's behalf • Perform actions that should be traceable • Enforce different permissions for different roles Here is an MCP-based web shop demo, built with WorkOS AuthKit and Next.js. With this example, you can do the following: • Connect to an MCP server • Authenticate using AuthKit • Order a T-shirt using MCP tools • View order details if you are an admin This GitHub repository contains everything you need: The WorkOS team has one of the best authentication platforms in the world. If you haven't looked into AuthKit, take a look at the code of this demo. Thanks to the WorkOS team for sponsoring this post.

Santiago

33,004 Aufrufe • vor 8 Monaten

Here is my submission for Vibecon by Emergent 1> I am an obssessive mission oriented builder. Give me something to accomplish and I will learn it on the fly. Built products > 120,000 users previously. 2> It would be great to work on intersection of chemistry and physics if I had unlimited budgets, to think from first principles. 3> Building Roger AI, a vision LLM tool which can guide you to accomplish any task!
3:10

Here is my submission for Vibecon by Emergent 1> I am an obssessive mission oriented builder. Give me something to accomplish and I will learn it on the fly. Built products > 120,000 users previously. 2> It would be great to work on intersection of chemistry and physics if I had unlimited budgets, to think from first principles. 3> Building Roger AI, a vision LLM tool which can guide you to accomplish any task!

Vaibhav Maheshwari

15,697 Aufrufe • vor 4 Monaten

I'm not a 3d modeler. I barely know how to open blender let alone do anything functional in it. But with the siddharth ahuja blender MCP and Claude I was able to vibe code an entire environment for my upcoming grow a garden game in Portals Claude used the MCP to assemble and arrange every single asset in this space (i literally don't even know how to import an asset into blender lol) within the spawn of a few hours and with access to a folder of low poly assets. When I had it start doing its own modeling, it even matched the style of the environment to keep things consistent If you are building UGC games and you don't know anything about modeling or design like me, you gotta try it out...
0:45

I'm not a 3d modeler. I barely know how to open blender let alone do anything functional in it. But with the siddharth ahuja blender MCP and Claude I was able to vibe code an entire environment for my upcoming grow a garden game in Portals Claude used the MCP to assemble and arrange every single asset in this space (i literally don't even know how to import an asset into blender lol) within the spawn of a few hours and with access to a folder of low poly assets. When I had it start doing its own modeling, it even matched the style of the environment to keep things consistent If you are building UGC games and you don't know anything about modeling or design like me, you gotta try it out...

Hugh MFER (PORTALS Guy)

180,406 Aufrufe • vor 6 Monaten

I built an agent that answers machine-learning questions. It's autonomous, and the best part is that I built the whole thing without writing a single line of Python code. Here is what I did and how I did it: Over a year ago, a friend and I built a site that publishes multi-choice questions. You get a new one every day. I decided to have GPT-3.5 answer questions. Here is what I needed to build: 1. Connect to the site's API to retrieve today's question 2. Extract the question and the potential choices 3. Connect to OpenAI's API and ask GPT-3.5 to answer the question 4. Parse the answer from the model 5. Submit the answer back to the API to get the score Not difficult. Likely several hours of work. But I didn't have to write any code. I built the whole thing by dragging and dropping components using Vellum is a YC-backed platform for developers to build LLM applications. They are the only ones I've seen offering this functionality. They sponsored this post, and their team helped me with all my questions while I built this. I created a workflow. The platform supports several node types to build whatever you have in mind. I show how I put the whole thing together in the attached video. The only code I had to write was a few lines of Jinja to parse and transform the API and the LLM results. There are three lessons I want to share from this experience: First, the best possible code is the one you didn't write. I'm a big fan of no-code tools because they help me materialize my ideas fast. They help product people, designers, and no coders collaborate on the solution. Second, Large Language Models are sensitive to how you prompt them. Small changes to prompts can make a big difference in results. This is more pronounced when you are building a multi-step workflow. Third, automated testing and evaluation for prompts is critical. There aren't many companies thinking about this. They'll have a hard time moving from a demo phase. The attached video will show you what I did.
3:46

I built an agent that answers machine-learning questions. It's autonomous, and the best part is that I built the whole thing without writing a single line of Python code. Here is what I did and how I did it: Over a year ago, a friend and I built a site that publishes multi-choice questions. You get a new one every day. I decided to have GPT-3.5 answer questions. Here is what I needed to build: 1. Connect to the site's API to retrieve today's question 2. Extract the question and the potential choices 3. Connect to OpenAI's API and ask GPT-3.5 to answer the question 4. Parse the answer from the model 5. Submit the answer back to the API to get the score Not difficult. Likely several hours of work. But I didn't have to write any code. I built the whole thing by dragging and dropping components using Vellum is a YC-backed platform for developers to build LLM applications. They are the only ones I've seen offering this functionality. They sponsored this post, and their team helped me with all my questions while I built this. I created a workflow. The platform supports several node types to build whatever you have in mind. I show how I put the whole thing together in the attached video. The only code I had to write was a few lines of Jinja to parse and transform the API and the LLM results. There are three lessons I want to share from this experience: First, the best possible code is the one you didn't write. I'm a big fan of no-code tools because they help me materialize my ideas fast. They help product people, designers, and no coders collaborate on the solution. Second, Large Language Models are sensitive to how you prompt them. Small changes to prompts can make a big difference in results. This is more pronounced when you are building a multi-step workflow. Third, automated testing and evaluation for prompts is critical. There aren't many companies thinking about this. They'll have a hard time moving from a demo phase. The attached video will show you what I did.

Santiago

309,825 Aufrufe • vor 2 Jahren

If your MCP server has dozens of tools, it’s probably built wrong. You need tools that are specific and clear for each use case—but you also can’t have too many. This creates an almost impossible tradeoff that most companies don’t know how to solve. That’s why I interviewed my friend Alex Rattray (Alex Rattray), the founder and CEO of Stainless. Stainless builds APIs, SDKs, and MCP servers for companies like OpenAI and Anthropic. Alex has spent years mastering how to make software talk to software, and he came on the show to share what he knows. I had him on Every 📧’s AI & I to talk about MCP and the future of the AI-native internet. We get into: • Design MCP servers to be lean and precise. Alex’s best practices for building reliable MCP servers start with keeping the toolset small, giving each tool a precise name and description, and minimizing the inputs and outputs the model has to handle. At Stainless, they also often add a JSON filter on top to strip out unnecessary data. • Make complex APIs manageable with dynamic mode. To solve the problem of how an AI figures out which tool to use in larger APIs, Stainless switches to “dynamic mode,” where the model gets only three tools: List the endpoints, pick one and learn about it, and then execute it. • MCP servers as business copilots. At Stainless, Alex uses MCP servers to connect tools like Notion and HubSpot, so he can ask questions like, “Which customers signed up last week?” The system queries multiple databases and returns a summary that would’ve otherwise taken multiple logins and searches. • Create a “brain” for your company with Claude Code. Alex built a shared company brain at Stainless by keeping Claude Code running on his system and asking it to save useful inputs—like customer feedback and SQL queries—into GitHub. Over time, this creates a curated archive his team can query easily. • The future of MCP is code execution. Instead of giving models hundreds of tools, Alex believes the most powerful setup will be a simple code execution tool and a doc search tool. The AI writes code against an API’s SDK, runs it on a server, and checks the docs when it gets stuck. This is a must-watch for anyone who wants to understand MCP—and learn how to use them as a competitive edge. Watch below! Timestamps: Introduction: 00:01:14 Why Alex likes running barefoot: 00:02:54 APIs and MCP, the connectors of the new internet: 00:05:09 Why MCP servers are hard to get right: 00:10:53 Design principles for reliable MCP servers: 00:20:07 Scaling MCP servers for large APIs: 00:23:50 Using MCP for business ops at Stainless: 00:25:14 Building a company brain with Claude Code: 00:28:12 Where MCP goes from here: 00:33:59 Alex’s take on the security model for MCP: 00:41:10
51:39

If your MCP server has dozens of tools, it’s probably built wrong. You need tools that are specific and clear for each use case—but you also can’t have too many. This creates an almost impossible tradeoff that most companies don’t know how to solve. That’s why I interviewed my friend Alex Rattray (Alex Rattray), the founder and CEO of Stainless. Stainless builds APIs, SDKs, and MCP servers for companies like OpenAI and Anthropic. Alex has spent years mastering how to make software talk to software, and he came on the show to share what he knows. I had him on Every 📧’s AI & I to talk about MCP and the future of the AI-native internet. We get into: • Design MCP servers to be lean and precise. Alex’s best practices for building reliable MCP servers start with keeping the toolset small, giving each tool a precise name and description, and minimizing the inputs and outputs the model has to handle. At Stainless, they also often add a JSON filter on top to strip out unnecessary data. • Make complex APIs manageable with dynamic mode. To solve the problem of how an AI figures out which tool to use in larger APIs, Stainless switches to “dynamic mode,” where the model gets only three tools: List the endpoints, pick one and learn about it, and then execute it. • MCP servers as business copilots. At Stainless, Alex uses MCP servers to connect tools like Notion and HubSpot, so he can ask questions like, “Which customers signed up last week?” The system queries multiple databases and returns a summary that would’ve otherwise taken multiple logins and searches. • Create a “brain” for your company with Claude Code. Alex built a shared company brain at Stainless by keeping Claude Code running on his system and asking it to save useful inputs—like customer feedback and SQL queries—into GitHub. Over time, this creates a curated archive his team can query easily. • The future of MCP is code execution. Instead of giving models hundreds of tools, Alex believes the most powerful setup will be a simple code execution tool and a doc search tool. The AI writes code against an API’s SDK, runs it on a server, and checks the docs when it gets stuck. This is a must-watch for anyone who wants to understand MCP—and learn how to use them as a competitive edge. Watch below! Timestamps: Introduction: 00:01:14 Why Alex likes running barefoot: 00:02:54 APIs and MCP, the connectors of the new internet: 00:05:09 Why MCP servers are hard to get right: 00:10:53 Design principles for reliable MCP servers: 00:20:07 Scaling MCP servers for large APIs: 00:23:50 Using MCP for business ops at Stainless: 00:25:14 Building a company brain with Claude Code: 00:28:12 Where MCP goes from here: 00:33:59 Alex’s take on the security model for MCP: 00:41:10

Dan Shipper 📧

15,645 Aufrufe • vor 10 Monaten

Raindrop cofounder ben hylak goes deeper on his exchange with Brian Chesky about booking Airbnbs with LLMs: "If Airbnb had an API, I would use it, and I would book Airbnbs through it using Claude Code. I know I would do it." "You see companies almost reducing themselves into an API with absolutely no moat. You look at Photoshop, Illustrator, etc. They're like, 'We have a Claude Code integration now.'" "At the point where people are just using Photoshop, Illustrator, etc. as an MCP, they've sort of lost the game, if no one is touching the UI anymore." "Right now, companies have to do that, increasingly, because they have no other choice. I think there will be a point where the incentives don't make sense anymore."
2:14

Raindrop cofounder ben hylak goes deeper on his exchange with Brian Chesky about booking Airbnbs with LLMs: "If Airbnb had an API, I would use it, and I would book Airbnbs through it using Claude Code. I know I would do it." "You see companies almost reducing themselves into an API with absolutely no moat. You look at Photoshop, Illustrator, etc. They're like, 'We have a Claude Code integration now.'" "At the point where people are just using Photoshop, Illustrator, etc. as an MCP, they've sort of lost the game, if no one is touching the UI anymore." "Right now, companies have to do that, increasingly, because they have no other choice. I think there will be a point where the incentives don't make sense anymore."

TBPN

109,350 Aufrufe • vor 2 Monaten

This is how I use LLM to scrape 99% of websites Many people didn't realise you can build agentic scraper to: 1. Handle Authentication, Human verification, Captcha 2. Handle pagination & complex UI interactions 3. Adaptive as website structure change 4. Scrape large set of data What used to take hours now can be automated in mins; Here I show case how do you automate a scraping job on Upwork where people are paying $50~$80 per hour; 0:00 Intro 1:54 Methods overview 4:52 Web Scraper agent using Spider Cloud 🕷️ Jina AI 8:37 Handle website auth & captcha using AgentQL 20:27 AI buy tickets MultiOn AI If you have any further question or want to get deep dive into the code example in the video, you can join my community where I post tips weekly:
22:43

This is how I use LLM to scrape 99% of websites Many people didn't realise you can build agentic scraper to: 1. Handle Authentication, Human verification, Captcha 2. Handle pagination & complex UI interactions 3. Adaptive as website structure change 4. Scrape large set of data What used to take hours now can be automated in mins; Here I show case how do you automate a scraping job on Upwork where people are paying $50~$80 per hour; 0:00 Intro 1:54 Methods overview 4:52 Web Scraper agent using Spider Cloud 🕷️ Jina AI 8:37 Handle website auth & captcha using AgentQL 20:27 AI buy tickets MultiOn AI If you have any further question or want to get deep dive into the code example in the video, you can join my community where I post tips weekly:

Jason Zhou

128,111 Aufrufe • vor 1 Jahr

MCP is an absolute game-changer. (Together with DeepSeek, MCP is probably the hottest thing in AI over the last 6 months.) I use Cursor to write code 90% of the time. I built an MCP server to connect the Cursor agent to GroundX, an open-source RAG system, and I'm not going back. This is officially insane! Here is what I did, step by step: First, a little bit of context. I maintain an end-to-end Machine Learning System with several pipelines to process data, train, evaluate, register, deploy, and monitor a model. I've written a lot of documentation explaining how the system works and how to modify and maintain it. There's also the documentation of the few libraries I used to build the system. I'm a massive fan of GroundX, an open-source enterprise-grade RAG system you can run on your servers or deploy to any cloud provider. I've been working with them for a long time. GroundX offers two services. First, the "ingest" service uses a custom, pretrained vision model to ingest and understand your data. I used this to process all the documentation I have for my code. Markdown files, source code, HTML files, and even PDF documents. Everything I've written related to my project went into GroundX. Their second service is "search," which combines text and vector search with a fine-tuned re-ranker model to retrieve information from the data. I needed to connect Cursor with this service, and that's where MCP came in. I built an MCP server with two tools: 1. The first tool would go to GroundX and retrieve the available topics. Splitting the data into topics (or "buckets," as GroundX calls them) allows me to use the same setup to serve documentation from different topics. 2. The second tool would search GroundX under a specific topic for the context related to the supplied query. The magic happens after connecting the MCP server with Cursor. Now, I can ask any questions related to my project, and Cursor's AI agent retrieves the list of available topics from the RAG system and then searches it to provide relevant context to the model. I went from getting mediocre, sometimes wrong answers to 100% truthful, complete answers. Here is the crazy part:
19:21

MCP is an absolute game-changer. (Together with DeepSeek, MCP is probably the hottest thing in AI over the last 6 months.) I use Cursor to write code 90% of the time. I built an MCP server to connect the Cursor agent to GroundX, an open-source RAG system, and I'm not going back. This is officially insane! Here is what I did, step by step: First, a little bit of context. I maintain an end-to-end Machine Learning System with several pipelines to process data, train, evaluate, register, deploy, and monitor a model. I've written a lot of documentation explaining how the system works and how to modify and maintain it. There's also the documentation of the few libraries I used to build the system. I'm a massive fan of GroundX, an open-source enterprise-grade RAG system you can run on your servers or deploy to any cloud provider. I've been working with them for a long time. GroundX offers two services. First, the "ingest" service uses a custom, pretrained vision model to ingest and understand your data. I used this to process all the documentation I have for my code. Markdown files, source code, HTML files, and even PDF documents. Everything I've written related to my project went into GroundX. Their second service is "search," which combines text and vector search with a fine-tuned re-ranker model to retrieve information from the data. I needed to connect Cursor with this service, and that's where MCP came in. I built an MCP server with two tools: 1. The first tool would go to GroundX and retrieve the available topics. Splitting the data into topics (or "buckets," as GroundX calls them) allows me to use the same setup to serve documentation from different topics. 2. The second tool would search GroundX under a specific topic for the context related to the supplied query. The magic happens after connecting the MCP server with Cursor. Now, I can ask any questions related to my project, and Cursor's AI agent retrieves the list of available topics from the RAG system and then searches it to provide relevant context to the model. I went from getting mediocre, sometimes wrong answers to 100% truthful, complete answers. Here is the crazy part:

Santiago

255,521 Aufrufe • vor 1 Jahr

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.
0:42

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.

Nick St. Pierre

377,975 Aufrufe • vor 2 Jahren

#emiamily TikTok live [260630] 🦊: I'm saving money right now, everyone. I want to drive. I miss driving. Actually, I love driving. I really loved it until I had to work at Asok. After that, I didn't like it anymore, bc the traffic is so bad, it really stressed me out. It took me two hours to drive from my home to Asok. Stressful! 🦊: But the other day, I went home and got to drive again and I was like, "I really want to drive again." But the air conditioning in the car at home is broken, so I got annoyed. When I had an event at a mall, I tried a test drive. I want to drive, but I don't really have knowledge. I think I need to gain more knowledge on this, bc I want to drive fast. But let the air conditioning in the car at home stay broken. The car is from 2013, it's 13 years old now. 🦊: It's not that I want to drive super fast, guys. It's just that I want to drive a car that's not a regular household car, do you get I mean? 🦊: A sports car might be too much. I don't have that kind of money 😆
2:18

#emiamily TikTok live [260630] 🦊: I'm saving money right now, everyone. I want to drive. I miss driving. Actually, I love driving. I really loved it until I had to work at Asok. After that, I didn't like it anymore, bc the traffic is so bad, it really stressed me out. It took me two hours to drive from my home to Asok. Stressful! 🦊: But the other day, I went home and got to drive again and I was like, "I really want to drive again." But the air conditioning in the car at home is broken, so I got annoyed. When I had an event at a mall, I tried a test drive. I want to drive, but I don't really have knowledge. I think I need to gain more knowledge on this, bc I want to drive fast. But let the air conditioning in the car at home stay broken. The car is from 2013, it's 13 years old now. 🦊: It's not that I want to drive super fast, guys. It's just that I want to drive a car that's not a regular household car, do you get I mean? 🦊: A sports car might be too much. I don't have that kind of money 😆

K-bab

54,125 Aufrufe • vor 1 Monat

Many of you are probably far too young to remember Lorne Greene's New Wilderness. This Canadian made CTV nature doc-series was wildly popular during its 5 year run from 1981-1986. In fact it was broadcast in over 20 countries including the United States where it was number one in its time slot. It was a show I grew up on - I watched it with my father. It framed how I see nature. It seems many full episodes are on YouTube. I urge you to check out as many as you can bc there's nothing like it today. For example watch this clip from the Great Horned Owl episode. [But note, you need to see the whole episode to grasp what I mean: I'd argue Lorne Greene's doc style 40 years ago is more powerful than any nature docs today. It doesn't have ANY of today's 5K bells and whistles. But the script writing...my God the writing...is spectacular. It hooks you in fast and before you know it, you can't stop watching and yearn for more. See for yourself: (also the original score symphony soundtrack is incredible!)
6:19

Many of you are probably far too young to remember Lorne Greene's New Wilderness. This Canadian made CTV nature doc-series was wildly popular during its 5 year run from 1981-1986. In fact it was broadcast in over 20 countries including the United States where it was number one in its time slot. It was a show I grew up on - I watched it with my father. It framed how I see nature. It seems many full episodes are on YouTube. I urge you to check out as many as you can bc there's nothing like it today. For example watch this clip from the Great Horned Owl episode. [But note, you need to see the whole episode to grasp what I mean: I'd argue Lorne Greene's doc style 40 years ago is more powerful than any nature docs today. It doesn't have ANY of today's 5K bells and whistles. But the script writing...my God the writing...is spectacular. It hooks you in fast and before you know it, you can't stop watching and yearn for more. See for yourself: (also the original score symphony soundtrack is incredible!)

Michael Gerald Gibbs🏳️‍🌈🍁 🇺🇦 (He/Him)

10,627 Aufrufe • vor 5 Monaten

Building AI agents and tools is getting ridiculously easy! Here is a pretty impressive FREE and Open-Source AI agent builder that runs on your computer. It's a desktop app you can install on Mac or Windows. Link: • Free • 100% no-code: You can type what you want, and the tool will build everything for you. • You can use local models like DeepSeek or Qwen, or connect to cloud models like GPT-4o, Claude, or Gemini with your API key. • You can turn the tools you build into MCP Servers. I recorded a quick video to show you how easy it is to build something from scratch.
4:10

Building AI agents and tools is getting ridiculously easy! Here is a pretty impressive FREE and Open-Source AI agent builder that runs on your computer. It's a desktop app you can install on Mac or Windows. Link: • Free • 100% no-code: You can type what you want, and the tool will build everything for you. • You can use local models like DeepSeek or Qwen, or connect to cloud models like GPT-4o, Claude, or Gemini with your API key. • You can turn the tools you build into MCP Servers. I recorded a quick video to show you how easy it is to build something from scratch.

Santiago

56,815 Aufrufe • vor 1 Jahr

Since Gemini 3 came out I haven't been able to stop thinking about the idea of personalized software - the idea that you can just make whatever app you need. Where I keep getting stuck is that it still makes me feel like the joy is in creation while consumption is still very much ephemeral. I keep making new things, but I don't persistently use any of them (nor do I really want to). Here's a good example: Gemini 3 fully one-shotted this poem explorer that gives you one poem a day, that explains each line at hover, and offers suggested questions to explore the theme further. It's pretty good! But I suspect I feel more joy and delight that it was able to make it vs. actually finding something like this... useful? Maybe the soonest, most effective type of personal software is games because they're comfortably transient - I can play for a bit, have fun, and keep trying other games (Roblox model) - and that can be the whole point.
0:45

Since Gemini 3 came out I haven't been able to stop thinking about the idea of personalized software - the idea that you can just make whatever app you need. Where I keep getting stuck is that it still makes me feel like the joy is in creation while consumption is still very much ephemeral. I keep making new things, but I don't persistently use any of them (nor do I really want to). Here's a good example: Gemini 3 fully one-shotted this poem explorer that gives you one poem a day, that explains each line at hover, and offers suggested questions to explore the theme further. It's pretty good! But I suspect I feel more joy and delight that it was able to make it vs. actually finding something like this... useful? Maybe the soonest, most effective type of personal software is games because they're comfortably transient - I can play for a bit, have fun, and keep trying other games (Roblox model) - and that can be the whole point.

Raiza Martin

103,868 Aufrufe • vor 8 Monaten

AI is making software development easier than ever. One day, maybe all SaaS will provide only an API key, with no user interface, and your AI will create the interface within your OS. Maybe it's a crazy idea, but it sounds possible (and awesome), so I gave it a try: I shipped 18 new API endpoints for my web analytics SaaS DataFast ✨ All your website data is accessible programmatically: unique visitors, total revenue, pages visited, countries, etc... That way, you can vibe code an entire analytics dashboard with a 1 prompt without worrying about the complexity of the backend! If you build something on top of it, please DM me. I'd love to check it out.
1:01

AI is making software development easier than ever. One day, maybe all SaaS will provide only an API key, with no user interface, and your AI will create the interface within your OS. Maybe it's a crazy idea, but it sounds possible (and awesome), so I gave it a try: I shipped 18 new API endpoints for my web analytics SaaS DataFast ✨ All your website data is accessible programmatically: unique visitors, total revenue, pages visited, countries, etc... That way, you can vibe code an entire analytics dashboard with a 1 prompt without worrying about the complexity of the backend! If you build something on top of it, please DM me. I'd love to check it out.

Marc Lou

134,884 Aufrufe • vor 6 Monaten

Deep Research is the most fun I've had with an AI tool since NotebookLM was launched! You can't imagine how incredible this agent feels for someone like me that loves doing research. I took it for a spin here and added my thoughts: I am impressed by the reasoning capabilities of the o3 model which operated under the hood. This is RL at its best once again! Synthesizing large amounts of information from different sources is no easy feat. I can tell you all about it from my experience building and scaling RAG systems. I have so many ways I plan to use this. More on this soon.
3:07

Deep Research is the most fun I've had with an AI tool since NotebookLM was launched! You can't imagine how incredible this agent feels for someone like me that loves doing research. I took it for a spin here and added my thoughts: I am impressed by the reasoning capabilities of the o3 model which operated under the hood. This is RL at its best once again! Synthesizing large amounts of information from different sources is no easy feat. I can tell you all about it from my experience building and scaling RAG systems. I have so many ways I plan to use this. More on this soon.

elvis

26,459 Aufrufe • vor 1 Jahr