Systems that survive load, failure, and time.
Senior backend/platform engineer (7+ years) focused on architecture that survives load, failure, and time. I lead design and delivery across multi-tenant SaaS, realtime connection planes, and cloud platforms on GCP and Cloudflare, with earlier AWS production ownership. Day-to-day toolkit: Kubernetes/GKE, Terraform, GitOps, PostgreSQL, Redis, TypeScript/Node.js, Laravel/PHP, observability, and high-volume HTTP/WebSocket systems.
Most recent production systems live in private company repositories. Public work below is reference engineering that proves the same decision quality without exposing employer or customer code.
| Signal | Result |
|---|---|
| HTTP plane | ~4M requests/day on multi-tenant education platforms |
| Realtime plane | 12M+ WebSocket connection sessions/day |
| Cloud migration | AWS → GCP production cutover with ~20 minutes controlled downtime |
| Leadership | Architecture direction, standards, reviews, mentoring, hiring |
| Repository | What it proves |
|---|---|
| spring-payments-platform | Java 17 / Spring Boot payments reference: idempotent transfers, immutable double-entry ledger, optimistic concurrency, transactional outbox, JWT scopes, signed webhooks, Testcontainers CI |
| realtime-platform-blueprint | Working WebSocket reference: tenant isolation, presence, backpressure, rate limits, Redis fan-out, metrics, load tests |
| production-platform-blueprints | Sanitized platform patterns: Terraform modules, GKE defaults, Argo CD GitOps, observability, incident/rollback runbooks |
| whatsapp-api-laravel | Production-oriented Laravel package for WhatsApp Cloud API integrations |
- Spring/PostgreSQL idempotency
- Outbox worker +
FOR UPDATE SKIP LOCKED - Optimistic concurrency + safe retry
- Spring Security JWT scopes
- HMAC webhook signing
- PostgreSQL immutable ledger constraints
| Area | Toolkit |
|---|---|
| Platforms & delivery | GKE, Cloud Run, Cloudflare Workers, Terraform, GitOps (Argo CD), private networking, Workload Identity |
| Realtime | WebSockets, Durable Objects patterns, Soketi/Redis fan-out, tenant isolation, backpressure |
| Reliability | Grafana / Prometheus / Loki, autoscaling, capacity planning, failure recovery, runbooks |
| Application & data | Laravel Octane, TypeScript/Node.js, PostgreSQL, Redis, Cloud SQL |
Designed and operated multi-tenant realtime infrastructure (Cloudflare Durable Objects and GKE/Soketi/Redis paths) with autoscaling, network isolation, independent deploy pipelines, and reconnect-friendly client contracts—supporting eight-figure daily session volume.
Architecture leadership for education SaaS operating at multi-million daily HTTP request volume: service boundaries, database capacity tactics, observability, and delivery safety.
Led production migration coordinating application, data, DNS, and infrastructure cutover with infrastructure-as-code, GitOps delivery, and end-to-end observability. Cutover measured in minutes (~20) for the controlled maintenance window.
- Architecture should make failure explicit and recovery boring.
- Observability is part of the design, not an afterthought.
- The best platform removes operational work from product teams.
- Performance, security, and cost are engineering constraints—not cleanup tasks.
- Public portfolio code is honest about scope: strong mechanisms, not fake product surface area.
Saudi Arabia · LinkedIn · GitHub · sudoasim.dev