I'm the kind of engineer who reads the RFC before writing the config. I invest in understanding how things work, then ship small, tested increments to validate what I learned. That foundation lets me iterate fast without breaking what matters.

• Scaling and securing AI infrastructure: capacity planning, service scaling, WAF security, and LLM prompt guardrails
• Improving CI/CD pipelines for LLM deployments: linting, integration testing
• Building agentic workflows for faster iteration: code review loops, cross-system investigations, automated test fixing

What I use on a daily basis

• Infra deployment: Terraform, CloudFormation, GitLab CI/CD, github workflows, feature branch deployments, integration testing, unit testing (Playwright, Vitest, Jest, Pytest)
• Cloud platform deployments
  • AWS services: CloudFront, ALB, ECS Fargate, EC2, VPC, Route53, S3, Lambda
  • Azure services: Traffic Manager, Application Gateway, API Management, Keyvault
• Production-ready AI infrastructure
  • Azure OpenAI, AWS Bedrock, Atlas Vector Search
  • API Management (load balancing, retry/fallback), WAF, private endpoints, Managed Identity
  • Datadog, Azure Monitor, custom LLM usage metrics
• Agentic workflows: Claude Code, Claude API
• System Analysis: Azure CLI, AWS CLI, Well-Architected Framework, AI-assisted exploration