Nghia Pham
Senior Full-Stack Engineer · AI/LLM Systems · React · Node.js
AI Product Engineer · Frontend Lead · Cloud & DevOps · 15+ years
Ho Chi Minh City, Vietnamxinchao@nghia-pham.comnghia-pham.comlinkedin.com/in/nia-d-phamgithub.com/llawliet11
Professional Summary
Senior Full-Stack Engineer with 15+ years shipping production web applications, now focused on AI-driven products: LLM-powered SaaS, RAG and retrieval infrastructure, and multi-agent systems. Built and operates a production AI memory platform (ContextQ), an AI-powered LMS with a live pilot customer (TutorOS), and a provider-agnostic agent-orchestration framework — on top of a decade of e-commerce, hospitality, advertising, and SaaS delivery. Equally strong on the frontend (React, Next.js, TypeScript) and backend (Node.js, Go, Python, PHP), with deep cloud-native experience (Cloudflare Workers, AWS, GCP, Kubernetes). Works AI-first across the whole lifecycle: LLM integration (Anthropic Claude, OpenAI, DeepSeek, Gemini), embeddings and hybrid vector search, MCP servers, and agentic development workflows.
Technical Skills
AI & LLM Engineering
LLM integration (Anthropic Claude, OpenAI, DeepSeek, Gemini, Cloudflare Workers AI), RAG & hybrid search (pgvector, Elasticsearch, FAISS, Vectorize), MCP (Model Context Protocol) servers, Multi-agent orchestration, Embeddings & semantic search, Prompt pipelines & LLM guardrails, Speech AI (Whisper STT, multi-provider TTS), AI-assisted SDLC (Claude Code, Codex)
Frontend
React, Next.js, Remix, Vue, Angular, TypeScript, JavaScript (ES6+), Tailwind CSS, HTML5/CSS3
Backend
Node.js, Express.js, Hono, PHP, Python, Go, REST APIs, GraphQL
CMS & E-Commerce
WordPress (themes, plugins, custom development), Hybris (SAP Commerce), Headless CMS
Cloud & DevOps
AWS, Cloudflare (Workers, Pages, D1, R2, Queues), Google Cloud Platform, Docker & Docker Swarm, Kubernetes, Prometheus & Grafana, CI/CD
Databases & Data
PostgreSQL (+ pgvector), TimescaleDB, MySQL, MongoDB, Redis, Elasticsearch, Apache Kafka
Selected AI Products
ContextQ — contextq.dev
2026 – Present
- Production multi-tenant memory and RAG platform for AI agents: hybrid semantic search combining PostgreSQL pgvector, Elasticsearch BM25, and RRF fusion, with pluggable embedding providers (OpenAI, Gemini, Voyage, Jina, Workers AI).
- LLM-judge pipelines for automated memory curation with multi-provider fallback; published MCP server plus JS and Python SDKs; evaluated on the LOCOMO and LongMemEval long-term-memory benchmarks.
TutorOS — tutoroshub.com
2026 – Present
- Multi-tenant LMS SaaS in production with a live pilot customer: LLM pipelines for question, distractor, rubric, and flashcard generation, automated open-answer grading, and an in-lesson Q&A assistant.
- Per-tenant AI audit logging and token accounting; Next.js, PostgreSQL/Drizzle, Auth.js, deployed on Cloudflare Workers with CI and 200+ test files.
OffloadCMS — offloadcms.com
2026 – Present
- Cloudflare-native CMS SaaS (Workers, D1, R2, Vectorize) replacing small WordPress sites with static-first, AI-search-optimized delivery; production control plane with CI-gated blue-green deploys and 200+ test files.
- LLM agent content editor with structured tool-calling (Claude via OpenRouter, Zod-validated edits only), Workers AI content generation, Vectorize semantic search, and credit-metered AI usage.
Cyberine
2026 – Present
- Provider-agnostic multi-agent orchestration framework (internal): adapters for Anthropic, OpenAI, AWS Bedrock, Google Vertex, and headless CLI agents; workflow DAG engine, parallel fleet dispatch, ELO-based model routing, and its own MCP server.
- Dogfooded daily to run parallel agent fleets across real client and product work.
Professional Experience
2026 – Present
- Embedded with a Vietnamese travel company as the hands-on engineer driving digital transformation across web, operations, and AI tooling.
- Rebuilt online booking flows (tours, airport fast-track, visa) on WordPress/WooCommerce: custom plugins, REST APIs, payment integration, rate-limited and hardened endpoints.
- Built a zero-touch CI/CD pipeline — merging a PR auto-deploys to production in seconds with byte-verified deploys and one-command rollback.
- Integrated bookings into Lark (Larksuite) CRM and rolled out guardrailed AI agent workflows so non-technical operations staff safely query and update the CRM — scoped permissions, allowlisted commands, full audit logging.
Senior Full-Stack Developer & Technical Consultant — Independent / Freelance Consulting
2010 – Ongoing
- Working AI-first since 2023 (early Anthropic/Claude adopter); since 2025, focused on AI-driven product engineering: production RAG and memory infrastructure, LLM content pipelines, speech AI, and multi-agent systems (see Selected AI Products above).
- Shipped further AI builds end-to-end: a Vietnamese-first TTS/STT SaaS orchestrating four speech providers plus Whisper and Llama 3.1 on Cloudflare Workers AI; an agentic shopping-research assistant with RAG (Vectorize + AI Search) and hallucination-guarding link verification; an API-discovery tool (trailcrumb.dev) with LLM suggestions and an MCP server.
- Built an LLM-powered operations copilot for a long-term industrial IoT client: Claude + OpenRouter with streaming chat, ~70 tool-calling functions over REST and telemetry APIs, prompt caching, token-budget guardrails, and rate limiting — engineered for horizontal scaling.
- Designed and operate the same client's AWS production infrastructure: Docker Swarm cluster, self-managed Kafka (KRaft), and a TimescaleDB time-series host with zero-downtime PgBouncer failover, monitored via Prometheus and Grafana.
- Delivered full-stack web applications and platforms to clients across e-commerce, agency, SaaS, and content domains, often running multiple engagements in parallel.
- Built React, Next.js, and Angular frontends; designed and shipped Node.js / Express APIs, integrations, and real-time Kafka data pipelines.
- Developed custom WordPress sites, themes, and plugins; built both traditional and headless setups.
- Operated cloud infrastructure on AWS, Google Cloud, and Cloudflare — networking, deployments, observability, and cost management — while providing technical consultancy and project leadership to remote, cross-functional teams.
May 2022 – Mar 2024
- Led the frontend team to deliver scalable, maintainable React-based web applications.
- Defined frontend architecture standards, performance budgets, and code-review practices.
- Collaborated with cross-functional teams to design, implement, and ship new features.
- Drove platform-wide performance, security, and SEO optimization; automated quality gates via CI/CD.
Senior Software Engineer (Full-Stack) — Vin HMS
Feb 2019 – Jun 2020
- Joined as full-stack engineer in an early-stage startup building a Hotel Booking & Management SaaS.
- Built backend services in Node.js, Express, and Apache Kafka for complex data flows and integrations.
- Developed responsive web applications with Angular and React for seamless user experiences.
- Deployed and operated cloud infrastructure on AWS for scale and reliability.
- Implemented end-to-end features for hotel booking, room management, and admin tooling.
Aug 2016 – Jan 2019
- Built and maintained frontend solutions on the Hybris (SAP Commerce) e-commerce platform.
- Delivered pixel-perfect, responsive UI in HTML5, CSS3, and JavaScript with strong cross-browser compatibility.
- Optimized frontend assets for performance and SEO across high-traffic storefronts.
- Built custom WordPress themes and plugins for client-specific needs alongside the primary frontend role.
Full-Stack Developer — Havas Riverorchid
Jan 2015 – Jul 2016
- Delivered full-stack web solutions for advertising campaigns at agency pace.
- Built campaign frontends with Angular; backed them with Node.js services.
- Ensured cross-browser compatibility, responsive design, and accessible deliverables.
Nov 2013 – Jan 2015
- Built websites end-to-end, owning frontend, backend, database design, and deployment.
- Maintained a portfolio of client websites for performance and uptime.
- Managed MySQL databases for data consistency and security across all platforms.
- Configured and operated servers ensuring reliable hosting.
Frontend / WordPress Developer — G.O.S.COM
May 2013 – Nov 2013
- Developed and maintained custom WordPress themes and plugins to client requirements.
- Performed cross-browser, multi-device testing of new features.
- Optimized site performance, security, and SEO; provided ongoing maintenance and troubleshooting.
Mar 2011 – Jun 2011
- Built and maintained responsive WordPress websites with rich multimedia content.
- Optimized pages for fast loading and collaborated with creative teams.
Education
Advanced Diploma in Information Technology
Singapore Informatics and Business Management Education — Grade: Excellent
Singapore Informatics and Business Management Education — Grade: Excellent
2010 – 2011
Diploma in Information Technology
Singapore Informatics and Business Management Education — Grade: Excellent
Singapore Informatics and Business Management Education — Grade: Excellent
2008 – 2010
Languages
Vietnamese — Native
English — Fluent — daily working language in a UK-based engineering team (CEFR ~C1)
Key Strengths
- AI-driven by default since 2023 — daily multi-agent development workflow (Claude Code, Codex, custom orchestration): 70+ custom agent skills, personal MCP servers, ~27B LLM tokens processed in the past month alone (~0.9B/day).
- Multi-role versatility — comfortable as IC, tech lead, or hands-on consultant.
- Multi-domain breadth — e-commerce, hospitality, advertising, SaaS, content, agency.
- Fast learner — quickly adapts to new stacks, codebases, and team contexts.
- Effective communicator — clear collaboration with engineers, PMs, designers, and clients.
- Ownership mindset — drives delivery from ambiguous requirement to production.
- Quality-focused — emphasis on performance, security, and maintainability.