VISHAL GUNJAL

Google Cloud Hackathon

Tech: FastAPI, GCP, Vertex AI, GitLab MCP, Terraform, Python, SSE

• The Problem: Infrastructure drift takes hours to patch manually, and traditional AI agents are too dangerous to automatically write and deploy Terraform directly into production.
• The Solution: Built an autonomous event-driven control plane using Google Vertex AI and GitLab that instantly detects misconfigured cloud resources and orchestrates the exact declarative Terraform code required for remediation.
• Production Safety (Blast Radius): Engineered a Python-based AST parser to statically analyze Terraform modules prior to execution, accurately calculating the deployment blast radius to guarantee zero automated breakages.
• Human-in-the-Loop GitOps: Designed an asynchronous pipeline leveraging Server-Sent Events (SSE) that streams proposed infrastructure changes in real-time, requiring explicit human approval before triggering CI/CD merges.
• The Metrics: Cut drift resolution time from hours to seconds, eliminated hardcoded credentials via Workload Identity Federation, and shifted FinOps from manual quarterly audits to continuous autonomous processes.

Tech: AWS EKS, Kubernetes, Terraform, ArgoCD, RabbitMQ, Redis, Prometheus, Grafana, Go

• Built an e-commerce platform with 8 independent microservices. Connected them with RabbitMQ so if one service crashes, the rest of the store stays online.
• Set up Redis caching to store frequently accessed product data, taking heavy read traffic off PostgreSQL databases.
• Locked down cluster security by writing Kubernetes Network Policies and using AWS IRSA so pods can access cloud resources without hardcoded passwords.
• Cut AWS setup time from 45 minutes to 15 minutes by writing reusable Terraform modules for VPCs and EKS clusters, saving state securely in S3.
• Reduced time to spot broken services to under 10 minutes using Prometheus, Grafana, and Jaeger distributed tracing.

Tech: AWS EKS, Jenkins, Terraform, SonarQube, Trivy, ArgoCD, ELK Stack, Docker

• Built a deployment pipeline using Jenkins, cutting execution time by 35% via parallel jobs and dropping Docker build times by 40% using layer caching.
• Set up automated Quality Gates that instantly block code from deploying if it has less than 80% test coverage or critical bugs.
• Integrated SonarQube, OWASP, and Trivy to scan code and containers, dropping production security issues by 95%.
• Automated deployments with ArgoCD, actively monitoring Git and instantly correcting infrastructure drift.
• Built Grafana dashboards to track engineering speed (DORA metrics) and system reliability (SLOs).