Exploring the world of OpenAI has been fascinating!
Recently, I put my learning to the test by incorporating the Text-Davinci-003 API into a project I've been working on - a simple book summarizer using Node and Next.js.
One of the most exciting aspects of working with OpenAI's models is the ability to engineer prompts for specific results. The more precise and thought-out the prompt, the more accurate and valuable the model's output.
Another interesting angle I explored was the token usage of the popular GPT-3 web interface.
Although it is currently free using the browser, when using the API endpoint in a project, the cost is calculated based on the characters of the prompt request and response. This added an extra layer of consideration in terms of balancing the operational cost of each request with the overall functionality of the project.
Overall, I really enjoyed building out this MVP and already have a list of features I would like to incorporate from feedback I've received.
- Understanding how to utilize the OpenAI API.
- Processing and manipulating text data efficiently
- Understanding the trade-offs between returning detailed book summeires and the operational costs of each API request.
- HTML5
- CSS3
- Next.js
- Node.js
- Open Ai API
Clone the repo as instructed below
No prerequisites
- Clone the repo
'https://github.com/CodeNameMoe/ai-book-review-app.git'
navigate to the front end folder "client"
Run: npm run dev
navigate to the back end folder "server"
Run: npm run start