IT 비전문가를 위한 AI 및 실무 도구 소개

TXI 클럽에서는 비전공자를 위한 생산성 향상과 실용적인 콘텐츠에 대한 새로운 책과 개발자를 위한 GitHub 기반 실습을 소개합니다.

### 신간 소개, 실용적인 콘텐츠, AI의 미래 트렌드
* 📚 연사는 GPT 모델 기반 앱으로 생산성을 **10배 향상**하는 데 초점을 맞춘 새 책을 소개하며 **PyTorch 및 TensorFlow API 통합**과 같은 고급 주제를 다룹니다.
* 💻 이 책에는 **Zapier 통합을 통한 앱 만들기, 다양한 검색 엔진을 GPT와 통합하기** 등 독자가 직접 따라할 수 있는 실습과 도구가 포함되어 있습니다.
* 🛒 이 책은 현재 **교보, 알라딘, 예스24, 인터파크에서 온라인으로 구매**할 수 있으며, 구매자가 간단한 후기를 작성하면 **온라인 강의에 참여**할 수 있는 이벤트도 진행 중입니다.
* 🤖 또한 챗봇, 멀티모달 콘텐츠 생성, 비즈니스 자동화 도구를 만들기 위한 **코드를 포함**하여 개발자를 위한 실용적인 연습 문제를 GitHub에서 공유합니다.
* 🌐 최근 라이브 세션에서 연사는 **실리콘밸리 OpenAI의 AI 및 휴먨노이드 로봇 개발, AI로 인한 빈부 격차 확대 가능성, 기본소득의 개념** 등을 언급하며 AI의 미래 트렌드에 대해 논의했습니다.
* 🎤 앞으로도 구독자들과 다양한 주제에 대해 토론하는 **라이브 세션을 정기적으로 개최**할 계획이며, 구독자들의 성원에 감사의 마음을 전합니다.
* 📚 연사 클럽인 XXI 클럽의 궁극적인 목표는 **비IT 전문가들이 AI를 효과적으로 이해하고 활용**할 수 있도록 돕는 것으로, 이들은 **AI 리터러시가 미래 생존의 핵심**이 될 것이라고 믿습니다.

Hello everyone, this is the TXI Club. Today's video is not about news or tutorial-related content, but rather I have a few things I personally want to introduce. I have a few things I want to introduce and promote, so I decided to film this video. First, I would like to briefly introduce a book that I have recently published. Second, I have been uploading videos mainly targeting non-major students, and I also have some practical content that I have created for them, so I will introduce that as well. Third, a few days ago, I did a live stream, but the sound quality was not good, so many people couldn't hear it well. This time, I have a microphone to deliver the content in better sound quality. First of all, my first book has finally been published. The screen is reversed so it might be a bit hard to see, but this is now showing GPTs.

I have now published a book on the topic of increasing productivity tenfold with an app. In the book, I cover various methods of creating the GPTS app based on the content of my existing videos. Primarily, I focus on more advanced topics such as PyTorch or TensorFlow API integration for tasks like business automation tools or search engine integration. In reality, fitting all the extensive content into just one video would take a long time, and I felt the video content was somewhat lacking, so I made an effort to include everything in the book. When you look at the table of contents, you will see that it is divided into three main parts. The first chapter introduces the basics of GPT and explains the fundamental operating principles, as well as providing a brief overview of prompt engineering. Since this book focuses on creating the GPTS app, it delves into details like prompt engineering.

Rather than just making apps, I introduce how to create practical tools that can actually be applied in work. In Chapter 2, I am now introducing various tools that can be created by integrating Zapier. This book focuses much more on practical exercises rather than theory. Instead of just discussing content, it is designed for readers to directly follow along, copy and paste prompts, and create apps themselves. By utilizing Zapier, you can create various apps such as scheduling assistants, automated customer service emails, data analysis, meeting summaries, automated thought organization, newsletter marketing, and more. Among the apps being created, some that can be integrated include Slack, Notion, Google Docs, LinkedIn, Gmail, and Excel sheets.

I am explaining about the method, and now in part 3, I will tell you how to integrate various search engines with GPT. Currently, we are using the ChatGPT and Bing search engines. If you want to integrate a search engine specialized for the Korean market, such as Naver, there are ways to do so. For example, you can create a Naver news automation, a YouTube recommended video bot, or even a bot that automatically finds job application forms and generates self-introductions. There are also methods for creating apps that automatically generate travel itineraries. Of course, I will share the prompts with you, and I also provide detailed explanations for the codes. You may not see it clearly, but I am sharing all the codes and explanations. Therefore, there are GitHub links available for direct access. Please make sure to refer to them.

Even if you don't have much knowledge, I solved a lot so that you can understand it. However, if there are still parts that you don't understand, please feel free to ask questions anytime through comments or an open chat room. Personally, it's a very proud book for me. We've been working with the publisher for almost six months to complete this book. Now, you can purchase it online at Kyobo, Aladdin, Yes24, and Interpark. Kyobo Bookstore, for example, has several stores where the book is available. If you are interested, please check the store locations and make a purchase. I would really appreciate it. Also, there is an ongoing event. If you purchase the book, authenticate it, and write a short review on Kyobo Bookstore or Yes24, you will have a chance to participate in the online lecture that I will conduct in June through a draw.

I am planning to offer nine lectures for free for twenty minutes. If you would like to participate in the certification event, please send a picture to my open KakaoTalk link, and I will complete your participation. Secondly, I have several practical exercises for developers that I would like to show you. These are currently available on my GitHub, and I have conducted several lectures on Lang game-related topics, so I have created some practice materials for that purpose. My GitHub is publicly available, divided into parts 1, 2, and 3. Starting with part 1, there are codes that explain how to use the basic Lang chain. Among the files, there is a file named 체포.py, which is a very simple basic example that can be run, and this is a code using the ChatGPT model.

When you actually deploy it, a chat window like this pops up, and you can create a chatbot that answers questions. In addition, there are a few other codes for different approaches. First of all, the upstage.py file is a arrest made using the upstage model. So, how does this look? This is the content I uploaded in the previous video. You can upload a PDF file here and there is a code prepared for a chatbot that can interact with Solr LLM. In addition, there are some .py files. This one uses the Rama framework, not the Lang chain framework, to utilize the Rama 3 and Py 3 models provided by Rama. So, now you can choose between the Py 3 and Rama 3 models for this LLM model, upload a PDF file, and ask questions about it.

I can create a lecture script, and I'm explaining about the code that can now create some basic rag reports. When you move on to part 2, part 2 is a section of code where you can do more creative things like multimodal images, voice, or vision. So, in the base image ipynb file, you can see codes to create different images like changing to 3 or to 2, and there are simple codes for recognizing vision images and for speech recognition (STT) and text-to-speech (TTS). Using these, you can make the codes more fun. For example, there is a multi-model py file that is only about 60 lines long. What this does is, when you upload an MP3 file with a link to a YouTube video, it does speech-to-text (STT) on it and creates something like SEO content articles on a website. So, let me give you an example.

I'll show you. If you upload a voice file, it will be converted into text, and if you actually play it, the White House has announced that it will take voluntary safety measures, such as inserting a watermark to prevent illegal duplication in AI-generated content by seven major AI companies like Alphabet and Meta OpenAI. Whisper is currently using this text to extract content and create SEO blogs that include SEO content based on this text. Therefore, it is a tool that generates various things such as summaries, titles, key words, subtitles, conclusions, etc., based on this content. It's a code book, and it's very good at English, so I hope you can also test it with an English voice file. So, the content for about 2 minutes is like this, and for about 3 minutes, it's related to business automation.

There are various codes, but first of all, there is a code that simply integrates email accounts in natural language. When asked to send an email to someone like L&M model, it can automatically send the email. Next, you can activate a code that utilizes the Serv API search engine to generate AI news from Naver newsletters, then request to transform it back into the Naver newsletter format. You can automatically send this content to your subscribers via email. Similarly, by creating an agent using Toolkit, you can ask it to send the newsletter content to a specific person, and it will automatically send it. Additionally, you can create automated agents for Slack bots. In the previous video, I discussed integrating Slack's features like this.

I have one video, and that is about the code for a Slackbot. If you want to add features like Google search or real-time web search to the Slackbot, you can simply copy and use the code called Tool Slackbot. There are a lot of codes for this, which are currently publicly available. It would be nice to try it out together. If there is more demand for this, I will consider creating a lecture on this topic. However, for now, the code is open source, so if you are curious, you can check it out here. Finally, what I talked about in the live session, there were many things to discuss, and to briefly summarize what I mainly talked about, OpenAI is gradually advancing, creating AI and developing humanoid robots in Silicon Valley.

People are really advocating for the advancement of technology rather than safety because they are true protego pro-advent enthusiasts. So, what will happen in the future is that the wealth gap may widen significantly, and it seems that only those who create and manage top 5% L&M models will be the ones to make a lot of money. I have thought and experimented on this idea, and if all jobs are replaced by AI in the next 10 or 20 years, what should those people do? Sam Altman is now creating a basic income called World Coin for that purpose. If we receive this basic income, we no longer need to work, and we start thinking about what to do with our lives. In that case, people will turn to entertainment, interests, or pleasures.

In industries like OnlyFans, it seems like people will end up paying more and more money. In fact, now people are somewhat like animals equipped with a system to work, so even if they receive basic income, how will people who want to work do? They might even start a rebellion. So now, Silicon Valley companies are creating metaverse virtual worlds, like VisionPro, so that if you want to work there, work. Create your own virtual world and live there. Because the future could be like this, I talked a bit about this. If such a world comes, will I enter that matrix? We also talked about this, and based on this, in the future, countries may need political systems. We were talking about such things, and later, AI boyfriends, AI girlfriends, and AI...

We have talked a lot about what to do in the future when self appears. It was late at night, but I decided to briefly go live to talk with the subscribers about it. I really enjoy doing this, so although it may be a bit challenging every week, I think it would be very fun to continue having conversations live occasionally in the future. Therefore, I will try that as well. Lastly, although I'm not sure how many people will watch this towards the end of the video, I would like to hold a book event. So, I am thinking it would be nice to have an offline book event with those who have purchased it. Maybe in about 2 weeks, on the following Saturday, we can meet offline, have some discussions, and network together. If you are interested, I will also conduct a survey.

I am planning to turn it, so if you could express your opinions there, I would appreciate it. So today, I have promoted various updates, and I would like to mention again that actually, I think I have been able to come this far thanks to my subscribers. If you hadn't shown such great interest, I wonder if I could have published a book like this. So, I really want to thank all the subscribers who made it possible for me to publish this book, and I will continue to bring a lot of content that will be even more helpful. The mission of the XXI Club is to help even non-IT professionals, even those who have no IT knowledge, make the best use of creativity. Therefore, I majored in computer engineering, but I also learned to code.

I started learning from college, so I understand how difficult it is to initially grasp coding or IT knowledge. I hope that everyone can utilize and understand this to the fullest extent. That's why I decided to create a club like this, aiming to minimize such hurdles and barriers as much as possible. My goal is to continue explaining things as easily as possible and help as much as I can. Because I believe that at this point, understanding AI will become crucial in the future. If the era I have been talking about in my live streams comes, only those who truly understand and utilize AI well may survive. That's why my ultimate goal is to make it understandable even for non-experts, ordinary people, my parents, grandparents, and even my great-grandparents. So once again, I am truly grateful for your interest so far. Thank you, and next.

I will come quickly through the video, so thank you for always watching. Thank you for today as well. Thank you.

나만의 GPTs 앱으로 생산성 10배 늘리기 | Langchain 무료 실습 자료, 구독자 감사 메세지

AI를 활용하고 싶은 개발자, 창업가들을 위해 모든 인사이트들을 공유해드립니다. OpenAI, GPT-4, LLM을 활용한 프로젝트들을 프로그래밍에 대해 아무것도 모르더라도 개발이 가능하도록 기초부터 차근차근 쉽게 설명해드립니다.

Learn everything about how to build OpenAI, GPT-4, LLM projects even if you don't know any programming at all. Will also be sharing insights for AI startups and B2B companies that want to use AI tools in their work. 


10X AI Club

새로운 프로젝트 소개 책, 실용적인 콘텐츠, 향상된 라이브스트림 품질

제작자가 GPTS 앱으로 생산성을 높이는 데 초점을 맞춘 새 책을 소개하며 PyTorch 및 TensorFlow API 통합과 같은 고급 주제를 다루고 있습니다.

이 책은 일정 관리 도우미, 자동화된 이메일, 데이터 분석 등 Zapier를 사용하여 다양한 도구를 만드는 실습을 강조합니다.

또한 네이버 뉴스 자동화, 여행 일정 생성 등의 작업을 위해 다양한 검색 엔진을 GPT와 통합하는 방법도 안내합니다.

또한 마이크를 사용해 콘텐츠의 음질을 개선하는 방법도 소개합니다.

신간 도서 출시 및 온라인 강의 이벤트

신간은 현재 교보, 알라딘, 예스24, 인터파크 등 다양한 온라인 서점에서 구매할 수 있습니다.

저자는 책을 인증하고 리뷰를 남긴 분들을 대상으로 추첨을 통해 온라인 강의에 참여할 수 있는 기회를 제공하고 있습니다.

또한, 저자는 깃허브에서 개발자를 위한 챗봇 제작 코드 예제, 랭 게임의 다양한 모델 작업 등 실습 예제를 공유하고 있습니다.

깃허브에는 멀티모달 이미지, 음성, 시각을 위한 창의적인 코드도 포함되어 있어 재미있고 혁신적인 프로젝트를 진행할 수 있습니다.

백악관, AI 생성 콘텐츠에 대한 자발적 안전 조치 발표

백악관은 주요 AI 기업의 워터마크 사용을 포함한 AI 생성 콘텐츠에 대한 자발적 안전 조치를 발표했습니다.

Whisper는 현재 이 텍스트를 사용하여 SEO 블로그를 비롯한 다양한 콘텐츠를 제작하고 있습니다.

이메일 계정 통합 및 Serv API 검색 엔진 사용을 위한 비즈니스 자동화 코드와 Slack 봇을 위한 자동화된 에이전트 생성도 논의되었습니다.

실리콘밸리에서 AI와 휴머노이드 로봇을 만드는 OpenAI의 발전도 간략하게 요약되었습니다.

기술 중심 세상에서 일과 소득의 미래

기술의 발전은 상위 5%의 L&M 모델 제작자를 중심으로 부의 격차를 확대할 수 있습니다.

샘 알트먼은 인공지능이 일자리를 대체하는 것에 대응하여 기본소득 이니셔티브인 월드 코인을 개발하고 있습니다.

사람들이 일 없는 삶을 고려함에 따라 엔터테인먼트와 가상 세계로의 잠재적 전환이 일어나고 있으며, 이는 새로운 정치 시스템의 필요성과 인간관계와 정체성의 미래에 대한 의문을 불러일으키고 있습니다.

구독자에 대한 감사와 XXI 클럽의 향후 목표

저자는 책 출간을 성원해 주신 구독자 여러분께 감사의 마음을 전하며 앞으로도 유용한 콘텐츠를 제공하겠다는 의지를 표명합니다.

XXI 클럽은 비전문가도 IT 지식에 접근할 수 있도록 하고 학습 장벽을 최소화하는 것을 목표로 합니다.

저자의 궁극적인 목표는 모든 연령과 배경을 가진 사람들이 AI를 이해할 수 있도록 하는 것이며, 미래에서 AI의 중요성을 강조하고 있습니다.