Inspiration
I started this project as a distributed URL shortener in C#/.NET because I wanted to get better at backend systems, networking, caching, reliability, and distributed-system design.
As I kept building it, I became more interested in a different question:
What if the system could help investigate its own failures instead of always waiting for an engineer to start from scratch?
That idea turned the URL shortener into a reliability-agent project. The goal became to build an agent that could inspect a real backend, remember previous incidents, use those memories during future investigations, recommend a safe action, and verify whether the action actually worked.
The project also became a practical way for me to learn how AI agents, persistent memory, backend observability, cloud deployment, and human-in-the-loop remediation can work together in one real system rather than as separate demos.
Log in or sign up for Devpost to join the conversation.