I updated the frontend in our explore page, deployed the frontend of our project, set up Github actions, and wrote frontend tests.
I attempted to deploy the backend with Emma but ran into unknown errors with the way our express was being installed. I also ran into typescript errors while using jest to run tests and had to edit our tsconfig file.
I'm planning on finishing up the profile page, creating the explore page, and helping my group members wrap up the project.
I mainly worked on implementing the OpenAI functionality this progress, which includes the lesson generation implementation w/ EH and creating the AI Helper chatbot with the create chat method from the API.
I mainly had issues implementing the OpenAI functionality for generating the lesson. EH and I tried to come up with the best prompt and cost effective way to call the OpenAI with consistent output as well as brainstorming about what we can do about the database design.
I'm planning on finishing up any missing functionality with OpenAI API, creating lesson finished page, and also helping my group members wrap up the project.
I helped work on the frontend for the user editing profile page as well as setting up the validation for this and the backend endpoints.
I also attempted to deploy the backend with EH and then troubleshoot on my own but couldn't solve the issue we were running into - thankfully EH and MW were able to reconstruct the backend and deploy the project.
I plan on helping finish the profile editing page, any frontend components I can help on and I will try to fix the navigation bar for the lessons to properly update for each respective lesson. I think we will also be spending time on ensuring we have proper documentation, comments and clean code and are ready for our demo.
I made the openai lesson generation endpoint with Daichi, and I restructured the database design. Made multiple endpoints like fetch + delete user, fetch lessons, fetch a single lesson, copy lesson, etc. I deployed the website with Maggie.
We had difficulties with openai lesson gen due to the complicated response structure and it was difficult to come up with a prompt that could be parsed correctly every time. After writing lots of complicated validation and parsing logic, we were able to get it to be mostly reliable. We also ran into unknown issues when deploying and had to spend a long time trying to get it to work
I will work on finishing all enpoints left in the backend