CustomAIzed

Home1
Home2
Inferences
Models

Inspiration

We have been inspired by generative models and their great capabilities for content generation. However, these tools are generally expensive, complex and have a high barrier to entry. We have set out to try to mitigate these barriers.

What it does

The platform allows users to enter images about a specific concept or person. After a training, they will have available an image generation model that has understood that presented concept, being able to play with the generative capabilities and the new knowledge of the model.

How we built it

For the machine learning tasks we used python with stable diffusion (diffusers). For the UI, we used Next.JS (node.js). Finally, to connect the architecture components and provide persistent storage to the system, we used supabase.

Challenges we ran into

Mainly we had to deal with new technologies for all team members, with all that this entails. We had never used supabase, tailwind and nextjs. As for the machine learning part, it should be noted that it is a very complex part that requires a lot of knowledge.

Accomplishments that we're proud of

The results are really good.

What we learned

We have learned a lot about service architecture and SaaS. The most positive aspect is that we have seen the great potential of supabase for future projects.