webapps and tutorials

This week has been exciting. I have this NLP project cooking inside my head (for some time now) and I’ve been speeding through many youtube tutorials on both backend and frontend structure.

For the backend, I’ve been cruising through Pinecone, LangChain and OpenAI API, thanks to tutorials from James Briggs and Data independent. Google colab has been my best friend.

Even though the backend is the new exciting stuff, I have a sweet spot for the way it’s presented. The frontend must:

I started with Flask/Dash vs Django, and after some thought, since I have built with it before, I picked Dash. I ended up building a scaffold, the development was faster this time around but it required some hardcode. The result was OK but not stunning. Dash is pretty taxing on layout and design. I deployed it on render (that was a first!)

Last night I discovered Streamlit. I was captivated by its beauty and simplicity and ended up creating a simple scaffolding with the help of Fanilo, Misra Turp and Coding Is Fun tutorials.

This morning I stumbled again on Pynecone (different pinecone) and I was amazed (thanks to NeuralNine’s video), it seem to have all I need for my simple implementation. It’s on alpha stage, but I guess I’ll give it a try.

If that wasn’t enough, it seems that FastAPI can also serve as a framework for a webapp. This will be fun…

I’m not sure what framework I will end up with. Let’s keep going!

Some stuff I found useful this week: