Adrian Liu

Bachelor’s degree & Computer Science

09/2010 – 09/2014

University of Edinburgh, Edinburgh, Scotland

Senior Full Stack Engineer

09/2022 – 02/2025

KPH Healthcare Services | East Syracuse, NY, United States

• Spearheaded end-to-end design of HIPAA-compliant patient management system for 23 departments, improving data entry speed, appointment scheduling, and communication efficiency, resulting in a 33% faster process (from 27 to 18 minutes) by Jan 2025
• Engineered React.js/Node.js modular architecture with role-based dashboards, achieving 32% faster user task completion (NPS 18% improvement) via dynamic form validation and real-time UI updates
• Scaled MongoDB cluster to 230k+ patient records through horizontal sharding (4-node replica set) by Dec 2024 and compound indexing, slashing read latency 40% (from 850ms to 510ms).
• Created zero-trust security model with JWT/OAuth2.0 flows, passing 15+ HIPAA audits (100% compliance rate) via cryptographic session management and HSM integration.
• Deployed WebSocket-driven notification system with 99.95% uptime, cutting critical lab result delivery time 20% through RabbitMQ message queuing and priority channels.
• Drove 50% page load improvement (Lighthouse 87+) via Next.js SSR hydration by 17th Nov 2024, implementing dynamic import/code splitting for core modules like EHR charting
• Streamlined CI/CD pipeline automation using Jenkins Groovy DSL, reducing deployment time by 40% (from 22 to 13 mins avg.) with parallelized testing/container builds.
• Containerized 12 microservices via Docker and Kubernetes (Helm charts), achieving 99.98% platform availability through pod auto-scaling and rolling update strategies.
• Maintained 98% Jest test coverage for 150+ React components, enforcing code quality via ESLint/Prettier hooks and SonarQube vulnerability scanning.

Full Stack Developer

06/2019 – 08/2022

Stepping Stone Advertising, New York, NY

• Developed an e-commerce platform with React/Node.js for a major retail company serving 175K+ customers by July 2022, boosting order throughput 13% via real-time inventory APIs and distributed transaction processing (Kafka event streaming).
• Implemented GraphQL for API communication, reducing the number of server calls by 28% and improving data fetching efficiency.
• Accelerated frontend performance from 5.2s to 2.3s through React.lazy dynamic imports, WebP image optimization, and Service Worker caching strategies.
• Integrated PCS-DSS compliant payment stack with Stripe/PayPal SDKs, achieving 10% higher transaction reliability via idempotent retry logic by Aug 2021 and 3D secure enforcement.
• Optimized AWS costs 20% ($48K/yr saved) using Lambda auto-scaling rules, S3 IntelligentTiering (1.2PB data), and CloudFront cache policies (92% hit rate).
• Enhanced security with mutual TLS, RBAC (Okta integration), and CSP headers – 40% fewer security incidents while maintaining 99.99% uptime SLA.

Full Stack Developer

01/2017 – 05/2019

Docebo | London, United Kingdom

• Built Next.js/React LMS platform scaling to 37K+ active users (25% growth), achieving 15% lower bounce rates through SSR implementation (SEO rank from 50th to 20th) and webpack code splitting.
• Integrated hybrid GraphQL/REST API layer (Django + Apollo Server), slashing data latency 20% while supporting 40% more concurrent users via query batching and Redis caching by Oct 2018.
• Modernized auth flows with JWT rotation/refresh tokens, reducing login duration from 4s to 0.9s by Feb 2019 and achieving Zero Truust compliance through HMAC-signed sessions.
• Drove 20% higher user retention via collaborative UX redesign – implementing automic design patterns and WebVitals monitoring (LCP > 1.2s across 95% of pages).

Junior Web Developer

10/2014 – 12/2016

GECKO Governance | Dundalk, County Louth, Ireland

• Architected Node.js event-driven platform for 50+ institutions, processing 2M+ transactions/day with 18% faster risk model execution via kafka-streamed market data pipelines.
• Developed real-time dashboards (D3.js/WebGL) achieving 92.3% reporting accuracy (+18.8% YoY) by Oct 2016 through schema-first data validation and Canvas GPU optimizations.
• Optimized MongoDB aggregation layers & Redis cache TTL policies, slashing query latency 40% (from 850ms to 510ms) via compound indexing and hash slot rebalancing.
• Institutionalized Jest/Mocha test automation with CI/CD guardrails, reducing critical bugs 20% through snapshot testing and contract-first API development.