Skip to content

Cloud DevOps & Full-Stack Engineer

Michael J. Loria

Building secureresilientautomatedscalableobservable cloud platforms.

I design, build, test, automate, secure and operate production grade multi cloud systems alongside robust, scalable and resilient containerized deployments using devops principles.

See my projects

About

About

Cloud engineer with expertise in AWS, Azure and GCP. I manage cloud infrastructure and develop full-stack software. Today I focus on designing multi-cloud infrastructure and building the full-stack apps that run on it.

AWS + Azure + GCP
AWS & Azure Certified · CLF-C02 & AZ-900
Bachelor of Science
Computer Science · CSULA

Certifications

  • AZ-900 Azure Fundamentals
  • CLF-C02 AWS Certified Cloud Practitioner

Projects and Experience

What I've built

Tap or click any tag for a quick definition.

  1. Project

    ShipTrack

    Multi-cloud migration from EC2 to Kubernetes

    • Planned and ran a cloud migration of a FastAPI shipment-tracking app from EC2 hosts to EKS across four repositories, assessing sixteen anti-patterns and choosing replatform, refactor, retain, or retire for each using the 7 Rs.
    • Moved traffic in waves by changing weighted ALB rules through pull requests, with a stack header on every response so a misrouted test fails. Rolling back is changing the weights back, and the legacy stack stays up as the fallback.
    • Took the same container image to AKS and to GKE Autopilot with Cloud SQL, Pub/Sub, and Workload Identity, using a cloud seam in the app and per-cloud Terraform.
    • Applied every infrastructure change through GitHub Actions over OIDC with a reviewer approving each apply, and a CI scrub that fails any pull request that commits an account ID or secret.
    View on GitHub
  2. Project

    CloudPunks

    Event-driven microservices marketplace on AWS EKS

    • Built a marketplace for 100 pixel-art characters as four microservices (product, inventory, order, and notification) behind a gateway, with a React UI and FastAPI services.
    • Connected the services through asynchronous events on EventBridge and SQS, using a transactional outbox so an order moves from pending to confirmed or rejected while inventory transfers ownership. Lambdas send low-stock and activity emails through SES.
    • Gave each service its own store (PostgreSQL, DynamoDB, and a Valkey cache), and provisioned the AWS environment with Terraform.
    • Ran the full system three ways: Docker Compose with LocalStack and no AWS account, a local Kubernetes cluster, and AWS EKS built and operated only by GitHub Actions.
    View on GitHub
  3. Project

    Beacon

    Three-tier incident tracker on AWS EC2

    • Built a service and incident tracking app with a FastAPI REST API, a React and TypeScript UI, and PostgreSQL, including an overview dashboard and a timeline on every incident.
    • Provisioned AWS with Terraform: a shared VPC with VPC endpoints and no NAT gateway, an Application Load Balancer, an Auto Scaling Group, and RDS PostgreSQL, running on a Packer-built base AMI with the CloudWatch Agent.
    • Automated delivery with GitHub Actions over OIDC (test, build, database migration, rolling instance refresh, and smoke tests), plus rollback to any commit, drilled with zero downtime.
    View on GitHub
  4. Project

    DORA Deployment Tracker

    Software delivery metrics platform

    • Built a three-tier app that records deployments, the commits they ship, and the failures they cause, and computes the five DORA software delivery metrics from that data.
    • Containerized the React, FastAPI, and PostgreSQL tiers with Docker Compose, using health-check-gated startup, least-privilege database roles, structured JSON logs, and Prometheus metrics.
    • Exposed an idempotent ingest API that CI pipelines call to report deployments. The app records its own builds through the same API.
    View on GitHub
  5. Project

    AWS CI/CD Framework

    Reusable multi-environment release pipeline

    • Built a reusable GitHub Actions deploy workflow that consumer repositories call by version tag, with dev, stage, and prod each building from their own branch behind required-reviewer approval.
    • Replaced long-lived AWS keys with GitHub OIDC federation. The deploy role trusts only approved environment identities, so an unapproved job cannot assume it.
    • Enforced container release discipline with multi-stage builds, non-root users, pinned base images, hadolint linting, and immutable SHA-tagged images, backed by a Terraform spec for the supporting AWS resources.
    View on GitHub
  6. Project

    Azure Hub-and-Spoke Network

    Azure networking lab

    • Deployed a three-VNet hub-and-spoke topology entirely from Azure CLI scripts, peering each spoke to the hub with no route between spokes so workloads stay isolated.
    • Locked down access with subnet-level NSGs: SSH reaches a single jump host only from the operator's IP, and spoke VMs are private and reachable only from the hub management subnet.
    • Wrote a verification suite that proves the isolation in both the control plane and the data plane, with deploy and destroy scripts that keep the lab cost disciplined.
    View on GitHub
  7. Education Los Angeles, CA

    California State University, Los Angeles

    Bachelor of Science in Computer Science

    • Systems and networking: Operating Systems, Internet Architecture, and Network Protocols.
    • Algorithms and security: Data Structures, Algorithms, and Cryptography.
    • Software and data: Software Engineering and Database Management.
    • Cloud and AI: Cloud Computing and Machine Learning.

Services

What I do

All repositories on GitHub

AWS, Azure and GCP Cloud Infrastructure

Containerized Deployment

Infrastructure as Code

Full-Stack Web Development

Monitoring

Security

Stack

Tools I work with

Cloud Platforms

Infrastructure as Code

CI/CD

Containers

Monitoring

Security

Scripting & Others

Full-Stack Development

Contact

Michael J. Loria

Platform Engineer