This article is the third edition of CryptoChasers' recommended high-quality Web3 educational resources, focusing on AI. AI large language models, represented by ChatGPT, have developed rapidly, and many experts believe that this will be a major trend in the next decade. Whether in our daily use, research, or development tools in Web3, we should learn and master the basics of AI to improve productivity.
There are many AI tutorials, and it is difficult to read them all. My recommendations are based on the advice of experts who have been working in this field for the past six months, combined with my own browsing experience.
Tutorial Recommendations#
wayToAGI#
- Website: wayToAGI
- Language: Chinese
- Type: Index
This website is an open-source knowledge space that gathers almost all useful information about AI. I highly recommend it and give it a 5-star rating. Here are some features I like about it:
- It collects and updates high-quality materials daily.
- It provides a learning path for AI beginners, prompts, AI drawing, AI voice, and an introductory guide to digital humans.
- It has various technical articles, both basic and in-depth, suitable for browsing and searching.
The following tutorials are for developers, preferably those who are familiar with Python or JavaScript.
OpenAI Official Documentation#
- Website: OpenAI Doc
- Language: English
- Type: Text
Although there are tools like LangChain that greatly simplify the development of LLM, it is still important to understand the principles behind these tools rather than relying on them completely. The best way is to read the official documentation and start learning from installing the official Python package with pip install openai or npm install openai. Some translations can be found in wayToAGI: ChatGPT Official Guide
ChatGPT Prompt Engineering for Developers#
- Website: ChatGPT Prompt Engineering for Developers
- Language: English
- Type: Video + Interactive
This introductory course, offered by Deeplearning.ai in collaboration with OpenAI, is taught by renowned educator Andrew Ng in the field of AI. It covers various types of prompt engineering, such as how to summarize and translate. The website provides online notebooks that allow you to run and debug code while watching the videos, without the need for environment configuration or API application. If you have difficulty understanding spoken English, you can watch the videos with Chinese subtitles.
Building Systems with the ChatGPT API#
- Website: Building Systems with the ChatGPT API
- Language: English
- Type: Video + Interactive
Also offered by Deeplearning.ai, this course is an advanced version of the previous ChatGPT Prompt Engineering for Developers course. It uses the development of a customer support assistant system as a case study to explain the principles and development process of ChatGPT. It includes some engineering content and can be complemented with OpenAI's GPT Best Practices.
LangChain for LLM Application Development#
- Website: LangChain for LLM Application Development
- Language: English
- Type: Video + Interactive
Also offered by Deeplearning.ai, this course is an introduction to LangChain, taught by the author of LangChain. LangChain is a framework for developing large language models (LLM), similar to React for developing frontend apps, which greatly simplifies LLM development. Going through the examples with the author makes it easier to get started compared to just reading the documentation. There is also another LangChain course called LangChain: Chat with Your Data, which focuses on how to use LangChain to create a local document question-answering tool, providing a practical approach.
What's Next?#
Next, I suggest that you develop your own tools based on LLM to deepen your understanding through practice. Although there may already be products on the market that meet your needs, developing a personalized tool that integrates with your workflow is undoubtedly a cool thing to do.
As for obtaining information about AI, it is sufficient to stay updated by following wayToAGI.
You can also try using other AI tools to see how others create AI products. I recommend the Web3+AI project MyShell, where holding their NFT allows you to use an API bot based on ChatGPT 4.0 for free up to a certain limit every day, and you can also earn NFTs by completing tasks.