Cloud platform architect
Cloud platforms.
Built to stay reliable.
I build cloud platforms that hold up under real production load, with reliability, security, and cost control designed in from the start. I also work directly with customers, in their environments, and build and run my own products end to end.
Latest case studies
Browse all case studies- An AI-assisted HRMS tool Leave, attendance, shifts, timesheets, and policy documents in one simple tool, with an assistant that quotes the handbook and the page. Read the case study
- Tolling S3 Migration: 1.48M Objects Moving 225 GB spread across 1.48M small objects with s5cmd on an existing in-region EC2 instance, at no extra compute cost. Read the case study
- Hardening a solo-built PDF product with AI How 10 automated test suites and a strict pre-push checklist kept AI-written code in check. Read the case study
Forward-deployed work
Close to the customer.
Through to production.
Working inside customer environments, building my own products, and tracing failures to their cause.
Discovery inside customer estates
Profiled 1,000+ servers across EU and US facilities, assessed migration readiness, and wrote the directional business cases used to qualify for AWS MAP funding.
A product built end to end
A PDF product taken from a broken serverless deployment to a self-audited, self-hosted app. First load is about 93% lighter and nine security findings are fixed.
An assistant that shows its sources
An HR tool whose assistant answers policy questions by quoting the company handbook with the document and page, and says so plainly when nothing matches.
Debugging live systems
Written root-cause analyses, from a silent call-record ingestion failure to an EKS ingress incident, tracing each from symptom to cause.
Expertise
How I work
From architecture to delivery.
The work connects architecture with delivery: setting up the foundation, moving workloads onto it, keeping costs in check, and making sure it stays healthy after launch. The learning paths explain these practices step by step.
Delivery process
How a change reaches production
Every change leaves a record of what changed, who approved it, and how to undo it.
-
Scope it
A work breakdown structure: work packages, owners, dependencies, and the order they land in.
Sequenced wave roadmap -
Raise the change
A change request with the risk, the rollback plan, and CAB approval before anything is scheduled.
Regulated CI/CD delivery -
Prove it
Review the plan output and rehearse outside production, so the real run holds no surprises.
Keep Aurora DR in parity -
Ship it
Roll out in stages inside the agreed window, with the rollback ready rather than improvised.
Rolling patches on Oracle RAC -
Verify and review
Check the outcome, not just that the deploy finished, then close it with a post-implementation review.
Silent CDR ingestion failure
Experience
A decade of delivery.
From systems to architecture.
I started in Linux systems engineering, moved into automation, and now lead cloud platform architecture across multi-account AWS environments.
Platform architecture in practiceDistributed storage engineering
Cloud Destinations · Lead Cloud Platform Architect
Lead architect for multi-account AWS governance, landing zones, database modernization, and cloud cost work for enterprise clients.
Spectromax Solutions · Senior Cloud Engineer
Worked on large database clusters and distributed storage behind ERP systems for manufacturing clients.
Vortex Technologies · Cloud/DevOps Engineer
Built the foundations: CI/CD automation, Linux administration, and early container work for Java apps on AWS.
Background & credentials
Certifications and education.
AWS Certified Solutions Architect, Associate
Amazon Web Services
HashiCorp Certified: Terraform Associate
HashiCorp
AWS Migration Acceleration Program (MAP), Accredited Practitioner
AWS Partner Program
B.E., Mechanical Engineering
KL University, 2010 to 2014
Get in touch
Questions about any
of this work?
I'm happy to talk through any case study or learning path: the architecture, the trade-offs, or why I made a particular call. Send a message here, or use email or LinkedIn.