I spent 2-3 days using ChatGPT 4o to optimize the UI for the website. I added a Ghibli-inspired style for the website colors and CSS. It looks much better! For the image generator part, I know many products only support uploading an image, then pressing a generate button and waiting. ChatGPT and Gemini use a chat mode to edit images. This time, I implemented a workflow where you generate an image, then press edit, and you can continue to adjust this image again - not in chat mode, which may be better. I also implemented asynchronous generation, so you don't need to wait for the image to generate - just add tasks to the tasklist, then add another image generation task. How's the design? I welcome your ideas and advice!