Part 0: Langchain AI app with Flutterflow & Flowise - No-code Crash Course Overview

how's it going Jasper from theose here welcome to this free course where we create a chat with PDF type of app with flood flow flow wise super base and open AI after this course you'll be able to create an app that your users can add their PDF documents into the knowledge base they can ask you to scrape a website and and add the scraped content into the same knowledge base then do a Q&A with the knowledge base and receive answers which is generated by Chad GP they can also go back to previous chat sessions and resume conversation with a particular PDF or website they have already added to their knowledge base I have my own chat with PDF or website app on the App Store so you can check it out to see what outcome you can achieve after this crash course this crash course does cover how flut tolow interface is put together how the database is designed and also how to make the action change on flutter flow but you will need to have a basic understanding of how flat to flow works I have some videos on my channel you can go check them out if you need more support you can always jump into my newly set up Discord channel to post your questions and I will usually reply within 24 hours you may also purchase these template app which is the exact app that I have published to the App Store you can purchase the flutter flow template to help you speed up your building process or just use it as a reference as you build up your own use cases however complete optional you don't have to purchase anything this course it's meant to be free and you can achieve the same outcome without purchasing the template app so the app is just there if you want to speed it up a little bit more but you definitely do not need to purchase the template enough said let's start this course and let's do a quick overview of why we're going with this particular tool stack and also how the tools work together from a holistic overview level why have we picked these tool stack the biggest benefit of these tool stack is your ownership to your tool stack and your data you also do not need to pay to get started with this tool stack all the required functionalities are available for free this tool stack protect you from vender locking risk because they are either open source or you can get the code after you have finished the project to host the anywhere that you want for example flow wise is open source and you can host it anywhere super base is open source and you can host it anywhere flutter flow the tool is not open source but you do own the entire codebase of your app once you have your app codes downloaded you can host your app or publish it anywhere you want without relying on flutter flow as the platform there are many ways to create knowledge based chatbot app uh without code for example there are many tutorials on YouTube already on how to create these type of apps with Bubble Plus flow wise plus pine cone however the biggest downside with that tool stack is that other than flow wise which is open sourced you are exposed to vendor locking risk also need pay plans for both bubble and pine cone to achieve the same results because some of the features we are going to be using is actually locked behind a pay wall so how do our tools work together to understand how the tools work together let's briefly cover what the user experience we're trying to achieve first users should be able to upload a PDF into the knowledge base users should be able to provide a website link and scrape the content into the knowledge base and users should be able to ask questions about the PDF uploaded or the website scraped and receive a response based on the knowledge and also based on their questions hence we can see that we first need our tool stack to work together to process the documents on the website and add that into our knowledge base and then have a way to ask questions and receive a response so let's talk about the add to knowledge based part which we can refer to as the upsert step users will interact with our flut tolow front end to upload their PDF or provide a URL and then we will send that PDF or URL with an API call to a specific flow wise endpoint which we will set up and the endpoint will work its magic to either load the PDF content or scrape the website content chop them up into chunks convert the chunks into embeddings which can then be upsert into our superbase Vector database an embedding is like a special encoding of all type of data into vectors you can see it as a way to format and store relationship between words so meanings and context are captured and beddings are formatted in a vector relationship so we need a vector database to store them a quick note here is that when we upser the embeddings into our Vector database we need a way to label them with metad dat so later when we want to search through them we can then provide the metadata as a filter so we are only searching through the embeddings with the desired metadata label instead of searching through every embeddings that we have added to our vector database once we have upseted our documental web content into our database we need a way to ask questions to retrieve question information and formulate answers we can call this asking a retrieve process a query users will first interact with the flat flow front end to ask a question through an API call to our flow wise endpoint which we specifically set up to handle queries our flow wise query chat chat flow would then retrieve information and generate a response we then serve the response back to our users as a message on the flutter flow front end now that we have gone through an overview of our tool stack the benefit and also how they work together holistically let's get started with making it [Music] happen [Music]

Channel: Nomo Codes - AI apps & automation with no-code

Views: 646

Keywords: nocode, softr, airtable, design, ux, ui, web design, web development, frontend, backend, lowcode, 3d, spline, product, saas, framer, webflow, googlesheet, zapier, make, integromat, bubble, figma, adobe, plugin, ios, android, pwa, sparkar, lensstudio, chatgpt, gpt4, openai, flutterflow, aigen, generativeai, generativedesign, nocodedesign, nocodeai, automation, MVP, startup, entrepreneur, ai, agi, artificial intelligence, generative ai, gpt, Langchain, Flowise, Chatwithpdf, Openai, Supabase

Id: TWCC1f7HXJU

Length: 6min 43sec (403 seconds)

Published: Mon Oct 30 2023