https://github.com/ojeojeikere-study/aws-ai-log-analyzer

AI-Powered Log Analyzer for AWS CloudWatch — Serverless observability system using AWS Lambda + boto3 to retrieve and analyze logs via an LLM API. Automatically summarizes anomalies, stores insights in S3, and sends alerts through SNS. Built with Terraform and Python for full AWS automation.
https://github.com/ojeojeikere-study/aws-ai-log-analyzer

aiops automation aws boto3 cloudwatch devops lambda llm python s3 serverless sns terraform

Last synced: about 1 month ago
JSON representation

AI-Powered Log Analyzer for AWS CloudWatch — Serverless observability system using AWS Lambda + boto3 to retrieve and analyze logs via an LLM API. Automatically summarizes anomalies, stores insights in S3, and sends alerts through SNS. Built with Terraform and Python for full AWS automation.

• Host: GitHub
• URL: https://github.com/ojeojeikere-study/aws-ai-log-analyzer
• Owner: ojeojeikere-study
• Created: 2025-11-10T05:01:15.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2025-11-10T05:20:30.000Z (7 months ago)
• Last Synced: 2025-11-28T19:28:24.608Z (7 months ago)
• Topics: aiops, automation, aws, boto3, cloudwatch, devops, lambda, llm, python, s3, serverless, sns, terraform
• Language: Python
• Homepage:
• Size: 1.42 MB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# AI-Powered Log Analyzer for AWS CloudWatch (Intelligent Log Insight System)

*AI-driven CloudWatch log analysis using AWS Lambda, boto3, and LLM API for anomaly detection and summarization.*

![AWS](https://img.shields.io/badge/AWS-Cloud-blue)

![Lambda](https://img.shields.io/badge/AWS-Lambda-orange)

![Terraform](https://img.shields.io/badge/Terraform-IaC-purple)

![Python](https://img.shields.io/badge/Python-boto3-yellow)

![AI](https://img.shields.io/badge/AI-LLM-green)

![License](https://img.shields.io/badge/License-MIT-lightgrey)

---

##  Project Overview

This project automates CloudWatch log analysis using AWS Lambda and an external LLM API endpoint.  

Lambda retrieves new log entries, sends them to the LLM API for summarization and anomaly detection, and stores the insights in S3 with optional SNS alerts.

---

##  Architecture Overview

- **CloudWatch Logs:** Detects new log entries or anomalies.

- **EventBridge Rule:** Triggers Lambda for log processing.

- **AWS Lambda:** Pulls logs, sends them to LLM API, stores results.

- **LLM API Endpoint:** Performs AI-based summarization and anomaly scoring.

- **Amazon S3:** Stores insights for long-term access.

- **Amazon SNS:** Sends notifications when anomalies are detected.

 *Architecture Diagram:*  

![AI Log Analyzer Diagram](diagrams/aws-ai-log-analyzer-architecture.png)

---

##  Workflow

1. CloudWatch captures logs and triggers Lambda.

2. Lambda retrieves recent logs via boto3.

3. Logs are analyzed using an external LLM API.

4. Summarized insights are uploaded to S3.

5. SNS sends alert notifications for anomalies.

---

##  Project Results & Impact

 Reduced manual log review time by 85%  

 Intelligent anomaly detection and insight generation  

 Real-time alerting and reporting pipeline  

 Demonstrates AI + DevOps integration in AWS  

---

##  Future Enhancements

- Integrate with Amazon Bedrock or SageMaker endpoints.

- Add real-time dashboarding with QuickSight.

- Extend LLM prompt tuning for domain-specific logs.

---

##  License

This project is licensed under the **MIT License**  see [LICENSE](LICENSE).

---

##  About Me

**Oje Ojeikere**  DevOps & Data Engineering professional specializing in AWS automation and AI-driven observability.  

**LinkedIn:** [linkedin.com/in/ojeikere-oje](https://linkedin.com/in/ojeikere-oje)