Harshil Yogi
AWS DevOps Engineer | Terraform, CI/CD & Automation
Profile in progress
Harshil is building their profile!
I built an AI-powered AWS log investigation system that allows teams to query logs stored in their existing S3 buckets using natural language.
The solution uses AWS Lambda, API Gateway, S3, Bedrock, and Terraform to:
- Convert natural-language questions into structured log searches
- Automatically discover and filter JSON, CSV, JSONL, and compressed log files
- Search across multiple S3 buckets without migrating log data
- Use Bedrock to summarize results and provide actionable insights
- Keep the log-search function private with IAM-based access control
The serverless architecture delivers typical queries in 2–4 seconds at approximately $0.001 per query, while keeping log data within the AWS environment.
Technologies:- AWS Lambda · API Gateway · S3 · AWS Bedrock · Terraform · IAM · Python
1
5
Automated Windows AMI Patching & AWS Infrastructure Automation
I built an automated AWS pipeline to eliminate the manual process of patching Windows servers and updating the AMI used by an Auto Scaling Group.
The solution uses AWS Image Builder, SSM Patch Manager, CloudWatch, EventBridge, Lambda, and CloudFormation to automate the complete patching lifecycle.
Automated Windows patch scanning and installation
Automated Sysprep and patched AMI creation
CloudWatch logging for patching and audit visibility
Automatic Launch Template and ASG updates
Preserved existing production configuration
The infrastructure is deployed through a single CloudFormation template, creating a repeatable and auditable workflow with minimal manual intervention.
My Role: AWS DevOps / Cloud Automation Engineer
Technologies: AWS Image Builder, SSM Patch Manager, EC2, Auto Scaling, Lambda, EventBridge, CloudWatch, CloudFormation, IAM, Windows Server, PowerShell
1
21
Designed and implemented an AI-powered AWS SRE automation platform to reduce manual infrastructure monitoring and security reporting.
The solution included two independent workflows:
• AI-powered EC2 monitoring that detects CPU usage patterns, analyzes them with AWS Bedrock, and generates scaling recommendations with an approval-based remediation workflow.
• Automated vulnerability reporting using AWS Inspector, Lambda, S3, Glue, Athena, Bedrock, and QuickSight to generate monthly security insights and executive dashboards.
Built the infrastructure using Terraform with an event-driven, serverless AWS architecture.
The solution improved infrastructure visibility, reduced manual operational effort, accelerated incident detection, and automated security reporting.
1
31