ringofchaos
Joined 2 years ago
Joined 2 years ago
The app will use genai to extract the details of houses listed for sale and then update my custom database.
For example I can input a youtube url and it will fetch the transcript and use llm to generate Json response based on predefined schema.
I can review and shortlist the houses based on various custom parameters using the web interface.
The mvp is done in a proprietary tech stack, I just need to port it to open source tech stack with React and FastAPI
The first part of converting data to csv works fine with help of ai coding assistant.
The reverse part of csv to database is getting challenging and even claude sonnet 3.7 is not able to escape newline correctly.
I am now implementation the data format in json which is much simpler.
I used it with React. My app is mostly financial data simulation and analysis and echarts fit the need
The documentation is great and with some tweaking you can make it look great. The customizing options are quite good.
I prefer it to some react only charting libraries .
There is a reason why python is used for ML for mathematical calculations. You could do all low level matrix multiplication using custom functions uisng any language. But is that optimal way to dob that.
My use case is similar
Your best option is to create backend api with python frameworks like Django Rest Frameworks or FastAPI. Frontend can be React which can be used for both web and mobile using React Native.
I had previous experience in python but had to learn React to handle similar requirements.
It's always best to use the right tool for right job.
I am actually satisfied that I went this route, added new skills and possibilities
You could also write native astro components which are very close to vanilla html css with added benifit of making reusable components and scoped css. And you can also use svelte or vue components.
I would also in future try to make it generic so that it can crawl any website and store new contents in vector databases. Response to user query then can be returned by combining the vector search and llm
But I still can't easily figure out to deploy my apps to a VPS.
So far I have manually set up my server and database.
I want learn more on devops to deploy it using containers.
This will enable me to host mutiple service in a VPS and I don't have to use PAAS solutions.
I can also easily switch providers too.