Hey everyone,
Over the past weeks I’ve been experimenting with FastAPI, caching strategies, fallbacks, and multi-source data aggregation.
To learn more about production-ready API design, I built two small developer-focused APIs:
🔹 1. Crypto Price API (multi-source)
This one aggregates prices for 26 major crypto assets, with fallback logic across:
CoinGecko
Binance
CryptoCompare
CoinPaprika
It includes lightweight caching and tries to avoid null values by returning the last known price if all sources fail.
Public API endpoint:
https://rapidapi.com/AstraScout/api/astrascout-crypto-api
(Main codebase is private, but the behaviour is fully documented in the public README.)
🔹 2. Market Insights API (Fear & Greed + market metrics)
A small aggregation layer that pulls:
Fear and Greed Index
Market conditions (dominance, volume)
Volatility
Momentum
Trending coins
A combined “insight score”
Public API endpoint:
https://rapidapi.com/AstraScout/api/astrascout-market-insights-api
(Again, the repository is private — I'm mostly testing design decisions and looking for feedback.)
🔍 What I'm looking for
Mainly opinions from developers who have built or worked with similar services:
Does the design/structure make sense?
Would you structure the fallback logic differently?
Tips for rate-limit handling and caching at scale?
Anything you would consider a “must add” for a production-ready version?
I’m not aiming to promote anything — just want to learn, improve, and eventually make cleaner real-world APIs.
Happy to hear thoughts from devs with more experience than me.
Thanks for reading!
— AstraScout
Top comments (0)