Backend and cloud engineer with 1.5 years of production experience. My ELT pipeline has processed 750,000+ records per nightly run, unattended, since August 2025. I've delivered 4 production systems for paying clients across logistics, finance, and healthcare — owning architecture decisions, AWS infrastructure, CI/CD, and post-launch support without a separate DevOps function.

Currently:

• Re-architecting a logistics platform from single-tenant to multi-tenant SaaS — row-level isolation, redesigned auth, billing integration
• Extending a government sector data lakehouse with a Neo4j knowledge graph for multi-hop traversal queries
• Migrating existing deployments from Elastic Beanstalk to ECS Fargate, provisioned via Terraform

End-to-end ELT pipeline for a government sector skills planning client — Azure SQL source → S3-backed raw, curated, and serving layers.

AWS Glue (PySpark)   DMS   Step Functions   S3   RDS   Neo4j   Node.js

• 750,000+ records ingested per nightly run, unattended since August 2025
• 400+ entities modelled in a Kimball-compliant dimensional schema with SCD Type 2; full point-in-time historical analysis
• Step Functions orchestration with conditional branching and automatic failure recovery across DMS, Glue, and downstream jobs
• Extending with a Neo4j knowledge graph for multi-hop entity traversal — replacing costly multi-join relational queries

Replaced a fully paper-based client operation. Led a 5-person team from requirements through production.

Node.js   PostgreSQL   React   Lambda   EventBridge

• Lambda + EventBridge automated month-end statement generation — eliminated a 2-day manual finance process
• Receivables aging report and subcontractor commission reporting built from scratch
• Currently re-architecting for multi-tenancy: row-level isolation with company_id across all tenant-scoped tables, middleware-enforced tenant context on every authenticated request

Prescription management, delivery route optimisation, and live driver tracking for a pharmacy operation.

Node.js   PostgreSQL   Prisma   AWS S3   AWS Location Services

• Nearest-neighbour route heuristic (O(n²)) sequencing 20–50 daily delivery stops per driver, with time-window overrides for priority runs
• Live driver tracking via 3-second polling architecture → PostgreSQL → AWS Location Services map
• Multi-layer prescription validation: dosage rules, controlled substance checks, and S3-backed document archiving for regulatory compliance

I write infrastructure as code. I instrument production systems before I consider them done. I document architecture decisions, not just implementations. I don't hand off to a DevOps team — I am the DevOps function.

Open to remote backend and cloud engineering roles globally.