Real-time API health dashboard showing requests, CPU, memory, latency, success rate, and incident monitoring.
Interactive analytics with CPU usage, memory monitoring, and endpoint latency visualization.
AI-powered assistant providing root cause analysis and optimization recommendations from API health metrics.
Slack AI Agent delivering API health reports and intelligent insights directly inside Slack.

API Guardian AI

Inspiration

Modern backend systems expose dozens of APIs, making it difficult for developers and DevOps teams to continuously monitor system health, identify performance bottlenecks, and respond quickly to incidents.

Most monitoring tools provide metrics but still require engineers to manually interpret dashboards and logs. I wanted to build an AI-powered assistant that not only monitors API health but also explains what is happening in plain English and provides actionable recommendations directly inside Slack.

This led to the creation of API Guardian AI.

What it does

API Guardian AI is an AI-powered API monitoring platform that combines a real-time analytics dashboard with a Slack AI Agent.

The platform continuously analyzes API metrics including:

Total Requests
Success Rate
CPU Usage
Memory Usage
Endpoint Latency
Error Counts
Incident Detection

The web dashboard visualizes system health through interactive charts and analytics.

Developers can also interact with the Slack AI Agent using natural language questions such as:

Why is CPU usage high?
Show API health summary
Which endpoint is slowest?
Give optimization recommendations

The AI analyzes the latest monitoring data and returns intelligent explanations instead of raw metrics.

How I built it

Backend

Node.js
Express.js
Groq LLM API
Slack Bolt SDK
REST APIs

Frontend

React
Vite
Tailwind CSS
Axios
Recharts

Deployment

Backend hosted on Render
Frontend hosted on Vercel

Challenges I ran into

One of the biggest challenges was integrating multiple systems together.

I had to connect:

React dashboard
Express backend
Slack Agent
Groq AI
Deployment across Render and Vercel

Another challenge was handling production deployment because frontend and backend were hosted separately. I had to debug API routing, CORS configuration, environment variables, and deployment issues before everything worked correctly.

Designing meaningful AI responses from monitoring metrics instead of returning generic text was another interesting challenge.

Accomplishments that I'm proud of

Built a complete AI monitoring platform from scratch.
Integrated an AI-powered Slack Agent.
Created a responsive analytics dashboard.
Implemented live API monitoring and incident reporting.
Successfully deployed the full stack application.
Learned how to combine AI with developer tooling to improve operational workflows.