Shorthills LogSense

• 

  Shorthills LogSense Architecture

Inspiration

Managing and debugging logs manually is time-consuming and error-prone. We wanted to automate log analysis with AI for faster insights and proactive issue detection.

What it does

Shorthills LogSense intelligently analyzes bulk logs, detects errors, suggests fixes, and sends alerts for critical issues. It also supports real-time monitoring via AWS CloudWatch.

How we built it

We built it using Python, Streamlit, and AWS Nova along with Strands Agents on Amazon Bedrock. The system leverages AI models for pattern recognition and error classification, seamlessly integrated with email alerts and an interactive dashboard.

Challenges we ran into

Handling large, unstructured logs and ensuring real-time analysis performance were key challenges. Integrating CloudWatch logs and managing asynchronous tasks also required careful optimization.

Accomplishments that we're proud of

We successfully automated log analysis, reduced manual effort, and created a clean, user-friendly dashboard for monitoring and insights.

What we learned

We learned AI-based error classification and building scalable monitoring systems using AWS Nova along with Strands Agents.

What's next for Shorthills LogSense

We plan to add predictive failure detection, multi-cloud support, and team collaboration features with automated incident reporting.

Built With

• amazon-bedrock
• amazon-cloud
• amazon-cloudwatch
• amazon-web-services
• aws-cloudwatch-ai/ml-services:-strands-agents-on-amazon-bedrock-databases:-(specify-if-you-used-any
• aws-nova
• dynamodb
• e.g.
• numpy
• numpy-platforms-&-cloud-services:-aws-nova
• pandas
• postgresql)-apis-&-integrations:-email-apis-for-alerts
• python
• streamlit