Microservices · MLOps

InfuraSaito

Go microservices that forecast cloud infrastructure load from live Prometheus metrics.

services behind one pipeline: Go + Pythonservices behind one pipeline

The problem

Capacity planning based on dashboards alone reacts too late. Forecasting load from the metrics already collected lets teams act before saturation.

What I built

1
Designed a Go microservices architecture around Prometheus metrics.
2
Served a Prophet forecasting model through a Python FastAPI service.
3
Containerized every service and automated tests with Docker and GitHub Actions.

Outcome

A tested, containerized forecasting service that plugs into an existing monitoring stack.

Have a similar problem?

Tell me about it. I reply within 48 hours with a first read of the problem and an honest opinion on fit.