Expert: loremipsum

A tool which is inspired by what lorem ipsum was built to create. Lorem ipsum was built to help designers focus on the design, and not on the content.....to focus on delivering the best UI/UX experience possible without designers troubling themselves with adding content....to focus on the bigger picture. Similarly, our tool, loremipsum allows designers to upload wireframes of their project and our tool converts it to code. It also converts code to make it compatible across platforms and frameworks, helping designers to spend time on what's important. Design. not functionality. this tool will generate code in multiple stacks and frameworks. So worrying about building codeable designs is in the past now.

i mean there are several assets which i didnt create.....(icons,images,mostly...) the animations were made by me...

Multiple challenges were encountered by me in my building process. I initially decided to use gpt4o-mini to convert screenshots(wireframes) to code but ran into API_KEY_NOT_FOUND errors. Then i tried integrating Llava but turns out it isnt really great at coding. I tried svelte but ran into a bunch of errors. The Huggingface library had multiple great models which i wanted to use; but eventually i found they were too big to run directly..and didnt support huggingspaces and inferenceapi since they were custom models. Together.ai doesnt support closed-source models and duplicate doesnt have much choice.

Then I found a starter guide at the IDX Project at Google which I used to then get started with Gemini. I refined the User Interface and finally got it to work. I looked for examples using threejs and integrated that. I learnt how to make a search bar animation with custom text on jitter.video. I did not know how to replicate the typewriter effect but GeeksForGeeks helped me. Thanks.