This is a simple website I made for my students where they can upload a .txt or PDF with C++ code in it and get an analysis of the runtime of the functions. It is a flask app that uses LangChain backend to pass the input through OpenAI.
The "Input" field displays the cleaned C++ functions within the file. This is because the file is no longer visible after being uploaded, and can contain other things besides C++ code (students can upload past exams which are several pages, but this would clean it down to leave just the C++ functions). The "Output" field is where the AI response message is printed.
An error handling feature exists where a browser alert message will pop up with any server side errors, to prevent the user from just seeing the loading bar spin indefinitely thinking it is working.
Current problems: 1 page PDFs cause an indexing error in the backend