### Purpose We have an idea on how technology can be used for learning and we want to explore it a bit further and run some experiments. If we can find conviction in the idea we can decide to pursue it further. ### Principles - Experiment - The idea is not to code out a perfect product or to constrain myself based on what we think we might be able to code out. The main goal is to get over the current mindset and get back into building things. ### Outcome Visioning Build out a quick and dirty pilot and start creating things again. Test it in one or two different settings with few users. Once we're feeling good about pursuing this we can work towards convincing others. ### Brainstorming - What do I already know about this project? I have a general idea of how I want to implement the idea but haven't worked out the details. I need to scope out an MVP and perhaps talk to someone with more of a teaching background. - What don’t I know that I need to find out? I need to brainstorm with a few others on how we can make the product look and what a MVP looks like. I particularly need to find the right topics and settings where this broader idea can be tested. - Who can I talk to who might provide insights? Nivedh might be a good person to talk to. Similarly can speak to my mom who also has considerable teaching experience. Harithas and Sobhana also might be good resources to reach out to. - What can I read or listen to for relevant ideas? I'm reading Seymour Papert already but I should also explore Jean Piaget and Resnick's ideas. ### General Actions - [x] Brainstorm with Harithas, Sobhana, Nivedh, Amma and Karthik ✅ 2024-02-29 - [x] Play around with GPT and see what kind of results you're getting ✅ 2024-03-12 - [x] Try separating out questions and see if the results can be improved ✅ 2024-03-13 - [x] Flesh out exactly what the product could look like ✅ 2024-03-12 - [x] Better documentation of all Git repositories ✅ 2024-04-04 ### AI Actions - [x] Replicate experiment for one topic in Chemistry ✅ 2024-03-20 - [x] Fine tune GPT 3.5 and experiment with it ✅ 2024-03-21 - [x] Create a test repository ✅ 2024-04-04 - [x] Set up all test cases ✅ 2024-04-04 ### Web App Actions - [x] Set up the T3 project ✅ 2024-03-15 - [x] Deploy T3 project ✅ 2024-03-15 - [x] Flesh out database schemas ✅ 2024-03-16 - [x] Set up Google Auth ✅ 2024-03-19 - [x] Build login screen ✅ 2024-03-19 - [x] Create database models ✅ 2024-03-23 - [x] Create logo ✅ 2024-03-22 - [x] Create colour palette and set up CSS ✅ 2024-03-22 - [x] Create home page with list of subjects ✅ 2024-03-22 - [x] Set up layouts properly ✅ 2024-03-22 - [x] Seed script for Database ✅ 2024-03-22 - [x] Build a list view with list of tests ✅ 2024-03-25 - [x] Create the main testing view ✅ 2024-03-28 - [x] Create test attempt on modal close ✅ 2024-03-28 - [x] Build endpoint to test explanation for each question ✅ 2024-03-29 - [x] Handle various question states ✅ 2024-03-29 - [x] Back and forward bug ✅ 2024-03-31 - [x] Connect web application with OpenAI assistants ✅ 2024-04-05 - [x] Fix text overflow in question explanation ✅ 2024-04-05 - [x] Clean up styles across the application ✅ 2024-04-05 - [x] Handle submit test show results somehow ✅ 2024-04-09 - [x] Make the instructions clearer ✅ 2024-04-09 - [x] Create OpenAI assistants for each test case ✅ 2024-04-17 - [x] Fix API timeout issue ✅ 2024-04-20 ### Potential Future Features - Set time limits on tests - Loading states for all api calls - Better logout page - Load test attempts and pick up where you left off - Investigate SSG for the pages - Prevent browser actions once test has started - Build data streaming capability so we don't run into timeout issues - Analytics and Tracking