Available for cloud / platform roles
Hi, I’m Abednego Tenge
Cloud Engineer AWS Certified Solutions Architect
I build secure, cost-efficient cloud infrastructure as code — and automate the boring parts so teams can ship faster.
Decorative terminal animation showing Terraform, AWS CLI and kubectl commands provisioning infrastructure.
01 — about
A bit about me
I’m a certified AWS Solutions Architect – Associate who learns by actually building. The projects below are real, working systems — designed, deployed and iterated on my own time, each one pushed to its own GitHub repo so you can see exactly how it’s built, not just a description of it.
I enjoy working at the intersection of cloud infrastructure, automation and software engineering, with a focus on designing systems that are not only functional, but resilient and maintainable. I’m particularly interested in cloud architecture, infrastructure as code, CI/CD, containers, monitoring, and finding better ways to automate repetitive processes.
I approach engineering with a mindset of learn, build, break, understand and improve. Whether I’m deploying infrastructure, troubleshooting an issue, or experimenting with a new cloud service, I enjoy understanding how things work beneath the surface and turning that knowledge into practical solutions.
I’m looking for a junior or associate cloud or platform role where I can bring that same hands-on rigor to a real team.
- AWS projects shipped
- 5
- AWS services used hands-on
- 20+
- Certifications
- 1
02 — skills
Tools I reach for
Grouped by what they actually do, not by how they look on a buzzword bingo card.
-
Cloud Platforms
-
Infrastructure as Code
-
Containers & Orchestration
-
CI/CD
-
Monitoring & Observability
-
Scripting & Languages
-
Security & Networking
03 — certifications
Certifications
Every badge links to its public verification page.
04 — projects
Things I’ve built
Five real AWS builds, each pushed to its own GitHub repo. The diagrams below are the actual architecture exports from each project — not stock icons.
-
Multi-Region Active-Active on AWS
Problem: a service running in one region goes down with that region, and most “multi-region” plans are diagrams that have never been tested against a real failure.
Solution: active-active across three AWS regions (us-east-1, us-west-2, eu-west-1), with Global Accelerator routing to the nearest healthy one and a deep health check that fails a region out of rotation if its DynamoDB Global Table replica goes stale, not just if it's unreachable. Verified with AWS FIS drills that kill a region's Lambda, network and replication all at once: 42–80s failover, zero data loss, and the recovering region held out of rotation (up to ~9 minutes) until its data caught back up.
-
ShopFlow — Event-Driven Order Platform
Problem: an order-processing backend that needs to survive an AZ failure without dropping requests, and hold up in a real design review.
Solution: CloudFront + WAF in front of an ALB → ECS Fargate → RDS Multi-AZ + ElastiCache on the synchronous path; EventBridge → SQS → Lambda (with a DLQ) on the async path. GitHub OIDC deploys with zero stored AWS keys. A forced RDS failover chaos test measured a 17-second app-level impact window.
-
Store Microservices Starter
Problem: show real service-ownership boundaries — not a monolith split into folders — where each service can deploy independently without touching another's data.
Solution: three independently deployable Node/Express services (Orders, Products, Inventory), each with its own Postgres database, running as separate ECS Fargate services behind one ALB across two AZs. Orders calls Inventory's internal endpoint to reserve stock before committing. Jenkins detects which service(s) changed via
git diffand only builds, pushes and redeploys those. -
Static Security Hosting
Problem: don't just flip on S3 static website hosting and point DNS at it — that's a public bucket, HTTP-capable, with zero request inspection in front of it.
Solution: a fully private S3 bucket readable only by one CloudFront distribution via Origin Access Control, AWS WAF (managed rule set, enforcing — not count-only) inspecting every request before it reaches CloudFront, an ACM-issued certificate, and WAF request logging (with the
Authorizationheader redacted) shipped to CloudWatch. -
SmartVault — Automated EC2 Backup & DR
Problem: an EC2 fleet with no backup story is one bad deploy or instance failure away from losing everything, with no disaster-recovery plan if the whole region goes down.
Solution: an EventBridge schedule runs a Lambda every 30 minutes that snapshots every EBS volume on the tagged Auto Scaling Group, copies each snapshot cross-region to us-east-2 for DR, prunes down to the 2 most recent per volume, and routes failures to SNS via a CloudWatch alarm and a Lambda destination — with X-Ray tracing on the whole run.
05 — experience
Where I’ve worked
My work history isn’t cloud-focused — the AWS work lives in the Projects section above, built on my own time.
-
Nov 2025 — Sep 2026
IT Support · Mother and Child Hospital
- Provided day-to-day IT support for hospital staff — troubleshooting hardware, software and network issues across clinical and administrative departments.
- Managed user accounts, access requests and device setup so staff could keep working without disruption.
- Documented recurring issues and fixes, building a reference base for faster resolution on repeat tickets.
06 — writing
Technical writing
Notes I wished existed when I was stuck.
07 — contact
Let’s talk
Hiring, contracting? Either box works.