Infrastructure Reliability · Cloud Security · Kubernetes · Infrastructure Automation
I build and investigate cloud infrastructure and platform systems where reliability, security, automation, and operational control intersect.
My work spans AWS, Azure, Kubernetes, Terraform, GitOps, observability, cloud security, and AI-assisted infrastructure operations.
A recurring problem across my work is the gap between:
what infrastructure is declared to be → what is actually running → what should happen when those states diverge.
That problem drives much of my work around infrastructure drift, security posture, Kubernetes operations, and controlled remediation.
Terraform drift detection → security analysis → cost impact → reviewable remediation
DriftGuard compares Terraform state against live AWS infrastructure and turns detected divergence into actionable engineering evidence.
- Detects infrastructure drift across supported AWS resources
- Maps findings against CIS AWS Benchmarks and MITRE ATT&CK
- Calculates estimated monthly cost deltas
- Generates Terraform HCL remediation
- Opens GitHub pull requests instead of modifying infrastructure directly
- Uses STS AssumeRole with workspace-specific external IDs
- Uses GitHub App installation tokens rather than shared PATs
- Includes CLI, API, dashboard and VS Code integration
AWS Terraform Python FastAPI PostgreSQL Redis GitHub Apps
AI-assisted infrastructure security analysis and controlled remediation
InfraGuard explores a harder problem than simply adding an LLM to DevOps:
How much authority should an AI system have when its output can change infrastructure?
The system:
- Processes infrastructure security findings
- Uses Azure AI Foundry and Phi-4 reasoning
- Performs structured root-cause and blast-radius analysis
- Applies a 70% confidence gate
- Generates Terraform remediation
- Creates GitHub pull requests
- Maintains an audit trail
- Uses OIDC / workload identity rather than static runtime credentials
- Never directly applies infrastructure changes
The final path is:
Finding → Analysis → Confidence Gate → Terraform → GitHub PR → Human Approval → Infrastructure
Azure AI Foundry Phi-4 FastAPI Terraform OIDC Kubernetes GitHub Actions
Repository · Live Dashboard · API
Kubernetes security posture + runtime detection + observability
A Kubernetes security pipeline combining:
- Trivy vulnerability scanning
- Kubescape MITRE ATT&CK assessment
- Falco runtime detection
- Prometheus metrics
- Grafana visualization
- ArgoCD GitOps delivery
- Terraform infrastructure provisioning
Example assessment recorded in the project:
80/100 Kubescape score · 101 Kubernetes resources assessed · 10 failed controls
Kubernetes Trivy Kubescape Falco Prometheus Grafana ArgoCD Terraform
Zero-trust Internal Developer Platform on Kubernetes
A GitOps-driven platform architecture built around:
- K3s
- ArgoCD
- Terraform
- Sealed Secrets
- Prometheus
- Loki
- GitHub Actions
The project focuses on the platform boundary between developers, workloads, infrastructure, secrets, deployment automation, and observability.
K3s Kubernetes ArgoCD Terraform Sealed Secrets Loki Prometheus
AWS · Azure
Kubernetes · K3s · ArgoCD · GitOps
Terraform · Helm · Kustomize · Ansible
IAM · OIDC · RBAC · NetworkPolicy · Trivy · Kubescape · Falco
Prometheus · Grafana · Loki · CloudWatch
Azure AI Foundry · Phi-4 · LangGraph
Python · FastAPI · Celery · Redis · PostgreSQL · Bash
Infrastructure Control
Platform Engineering
Kubernetes Reliability
Cloud Security
Infrastructure as Code
GitOps
Observability
AI-assisted Infrastructure Operations
I write about infrastructure problems rather than technology checklists:
- Cloud architecture and failure modes
- Kubernetes security and operations
- Terraform and infrastructure drift
- GitOps and platform engineering
- Reliability and observability
- Controlled automation of infrastructure operations
LinkedIn: Edwin Jonathan
Hashnode: @jonathandevops
GitHub: EdwinJdevops