Our inspiration for creating Net-sentinel came from realizing that very few AI tools truly support network engineers in their day‑to‑day tasks, and even fewer can help regular users who just need guidance with routing or connectivity issues. We set out to build a smart helper that understands what someone wants, figures out the steps required, performs the work safely, checks that everything is functioning correctly, and always keeps a human in the loop. To achieve this, we combined an AI agent capable of interpreting natural‑language requests with a set of specialized tools that execute tasks and report results back through a simple dashboard interface. Along the way, we faced challenges such as designing safe automation guardrails, handling vague or incomplete instructions, and balancing a clean UI with the technical depth network engineers expect. Despite these hurdles, we’re proud that Net-sentinel can already break down real networking tasks, execute multi‑step workflows, and validate changes to prevent misconfigurations. Building this system taught us how powerful AI becomes when paired with thoughtful UX, strong safety boundaries, and clear human oversight. Looking ahead, we plan to expand the toolset, integrate with more vendors and cloud platforms, enhance the agent’s reasoning for complex workflows, and introduce collaborative features so teams can review and approve changes together
Built With
- adobe
- api
- fastapi
- html
- postgresql
- python
- react
- squlite
Log in or sign up for Devpost to join the conversation.