Yeah having a serverless application AND a serverless db sounds great but then you realise you have two cold starts to deal with and god knows what else.
Still going. Recently built an endpoint to store and retrieve user profile info split between Cognito and my application. Was a bit ugly because there are some profile attributes that have to be unique.
First feature almost complete. I'm building out the API to be consumed by the react front end.
It's going well. One thing I'm finding tricky is getting dates in a not normal string format and then having to put them in date order for the response 🙃
Starting to finish off the back end functionality for one of the graphs. It'll be a dashboard heavy app so I'm going to take it one graph, then one page, at a time, trying to finish each feature end to end before moving on
Aside from that, I've been cleaning up the codebase because it is still really messy from when I was running EC2/SQLite. Need to change some basic bits on all the endpoints to account for the Aurora Data API. #indiehacker
Changed the architecture already 🙃 moved away from EC2 and SQLite to a serverless API and Amazon Aurora for the DB, so I never have to worry about scale and I can make use of the Aurora V1 data API.
Working on authentication with Cognito. Because this app is using a react.js front end, it's client rendered vs server rendered so totally different to what I've built before.
I'm #buildinginpublic starting today. Spinning up a brand new product to provide insight, tracking and analysis for track and field athletes' performance in the UK.
Pretty niche, but it's a big enough audience I think