Fahad Mushtaq, Senior AI Full-Stack Engineer: RAG assistants, LLM guardrails and evaluation, and the React and Node.js products around them
Senior AI Full-Stack Engineer
Building production LLM features — RAG assistants, guardrails and evaluation, computer-vision labeling and n8n automation — and the React and Node.js products around them.
Reason from what's in front of you.
Check every step.
This cube is scrambled at random and solved live in your browser, with nothing pre-recorded: 43 quintillion possible states and always a way through. It's the discipline I bring to production AI: ground every answer in the data actually in front of the model, verify each step, and fall back safely when confidence is low.
Works from the cube's current state, one stage at a time, the way a RAG assistant answers from retrieved sources instead of from memory. It solves a cube it has never seen.
- Planning…
Every scramble has a solution. Bring me the tangled one: a RAG assistant, AI guardrails, a labeling workflow, an automation, or the full-stack product around them.
AI features that hold up in production.
I'm a Senior AI Full-Stack Engineer with 10+ years of software engineering experience, building production LLM features for healthcare. I deliver RAG assistants and real-time AI summarization for a value-based care platform used by clinicians and care teams at health systems and ACOs, along with computer-vision data labeling workflows and n8n automation.
I work end to end, from AI pipelines and REST APIs to production React and Next.js web applications. Over 10+ years that has meant full-stack delivery at Cyber Advance Solutions, leading full-stack development of a core healthcare platform at Persivia, AI contracts like PAVE and Levers.health, and three DeepLearning.AI courses: Agentic AI, Advanced AI, and Safe and Reliable AI via Guardrails.
▋Ten years of tools, in orbit.
Pick a discipline to spin up the stack — hover the constellation to set it in motion.
- Generative AI & LLMs: RAG pipelines, Embeddings, Semantic search, Chunking, LLM integration, Prompt engineering, Grounding, LLM summarization, Conversational AI, LLM evaluation and monitoring, AI guardrails and governance
- AI Tools & Frameworks: OpenAI API, Anthropic Claude API, Google Gemini, LangChain, LangGraph, Hugging Face, Ollama, Vector databases (Qdrant), Python, FastAPI, OpenCV
- Agentic & Real-time Systems: Agentic workflow automation (n8n), Real-time AI streaming (WebSockets), Enterprise API integration
- ML Data Operations: Computer vision, Human-in-the-loop labeling, Label QA, Dataset curation
- Full-Stack Engineering: React, TypeScript, Next.js, JavaScript, HTML, CSS, Tailwind CSS, Bootstrap, Kendo UI, Node.js, Express, MySQL, REST APIs, Backend services, Reusable component architecture, Legacy platform modernization, Shopify
- Dev Tools & Platforms: Git, GitHub, GitHub Actions, Docker, PostgreSQL, MongoDB, Redis, React Native, Flutter, GoHighLevel (GHL), Redux, Vite, Postman, Swagger / OpenAPI, Jira, Vercel
- Leadership: Team mentoring, Code reviews, Engineering standards, Cross-functional delivery
Ten years of shipping, now with AI in production.
Senior AI Full-Stack Engineer
Persivia
Production LLM features for a value-based care platform used by clinicians and care teams at health systems and ACOs.
- Built a RAG clinical assistant in the Patient CMS that gives physicians source-cited answers using chunking, embeddings and semantic search
- Implemented AI safety guardrails: input/output validation, sensitive-data protection, grounded-answer checks and low-confidence fallbacks
- Set up LLM evaluation and monitoring to track answer relevance, faithfulness and accuracy in production
- Built a real-time Jira escalation dashboard that uses LLM summarization to surface critical items, SLA risks and owners, with live updates over WebSockets
- Built clinical dashboards that turn patient and care-management data into actionable views for clinicians and care teams
- Built and maintained a reusable React component library adopted across multiple modules
- Lead full-stack development of the core healthcare platform, including frontend modules, backend REST API integrations and legacy modernization; mentor the team through code reviews
Production LLM features for a value-based care platform used by clinicians and care teams at health systems and ACOs.
AI & Full-Stack Engineer
PAVE — AI Pavement Evaluation Platform
Computer-vision defect detection on road-inspection imagery, with engineers reviewing every model label.
- Integrated computer-vision ML models to detect and classify pavement defects (cracks, potholes, surface distress) in road-inspection imagery and auto-generate pre-labels
- Built a human-in-the-loop labeling workflow where engineers review, correct and approve model-generated labels
- Built QA review and dataset-management flows to keep label quality consistent and organize labeled data for model training
Computer-vision defect detection on road-inspection imagery, with engineers reviewing every model label.
Automation & Shopify Engineer
Levers.health
n8n automation for a direct-to-consumer telehealth brand across commerce, subscription and review platforms.
- Built n8n automation workflows syncing orders, subscriptions and customer data across commerce, subscription and review platforms, and triggering lifecycle notifications
- Migrated the recurring-subscription system to a new provider and rebuilt the reviews integration, preserving customer data and billing continuity
- Rebuilt the Shopify storefront and homepage
n8n automation for a direct-to-consumer telehealth brand across commerce, subscription and review platforms.
Full-Stack Developer
Cyber Advance Solutions
Full-stack web applications end to end, from requirements through deployment.
- Delivered full-stack web applications end to end: React frontends, Node.js/Express REST APIs and MySQL data layers, from requirements through deployment
- Improved application performance across the stack by optimizing front-end assets and refining API and database interactions
Full-stack web applications end to end, from requirements through deployment.
AI certifications
DeepLearning.AI
Agentic AI, Advanced AI, and Safe and Reliable AI via Guardrails.
- Agentic AI
- Advanced AI
- Safe and Reliable AI via Guardrails
Agentic AI, Advanced AI, and Safe and Reliable AI via Guardrails.
Bachelor of Science in Information Technology (BSIT)
Punjab University College of Information Technology
Bachelor of Science in Information Technology.
Bachelor of Science in Information Technology.
Currently open to new opportunities
Have a product to build or a team to strengthen? Let's talk.
Let's TalkWays we can work together.
RAG Assistants & LLM Features
Retrieval pipelines with chunking, embeddings and semantic search, grounded answers that cite their sources, and conversational AI built into your product.
Start a projectAI Guardrails & Evaluation
Input/output validation, sensitive-data protection, grounded-answer checks and low-confidence fallbacks, plus monitoring for relevance, faithfulness and accuracy.
Start a projectWorkflow Automation (n8n)
n8n workflows that keep your commerce, subscription and internal tools in sync and trigger the right notifications at the right time.
Start a projectFull-Stack Web Applications
React / Next.js frontends, Node.js / Express REST APIs and MySQL data layers, from requirements through deployment.
Start a projectComputer-Vision Labeling Workflows
Model pre-labeling, human-in-the-loop review, label QA and dataset management for ML training data.
Start a projectReal-Time AI Dashboards
Live dashboards over WebSockets with LLM summaries that surface what's critical, what's at risk and who owns it.
Start a projectNot sure which fits? Let's scope it together.
Get in touchSelected work, built end to end.
RAG clinical assistant and platform modernization
Persivia · Senior AI Full-Stack Engineer
A RAG chatbot inside the Patient CMS that gives physicians source-cited answers, with guardrails and production evaluation, plus a large-scale upgrade of the platform's legacy Kendo UI, Bootstrap and jQuery front end.
PAVE: AI pavement evaluation platform
PAVE · AI & Full-Stack Engineer
An AI SaaS platform where computer-vision models detect and classify pavement defects in road-inspection imagery and pre-label them, and engineers review, correct and approve every label before it becomes training data.
Levers.health: telehealth commerce automation
Levers.health · Automation & Shopify Engineer
n8n automation that keeps orders, subscriptions and customer data in sync across commerce, subscription and review platforms, a subscription-provider migration, and a rebuilt Shopify storefront for doctor-supervised peptide, weight-loss and longevity treatments.
ZipCoin Remit: cross-border money transfers
ZipCoin Remit · Full-Stack Engineer
One product across web and mobile for money transfers to 180+ countries: a 20+ page React web app with dashboards, and a full Flutter mobile app.
Divina Boutique: Shopify, GoHighLevel and n8n
Divina Boutique · Full-Stack & Automation Engineer
A luxury Shopify storefront for an Italian fashion boutique, connected to GoHighLevel (GHL) through n8n so every customer and order lands in the CRM and follow-ups run automatically.
FOEx: field inspection app
FOEx · React Native Developer
A React Native app for explosion-proof equipment inspections: structured checklists, QR scanning, reference material and real-time team chat in the field.
youPark: real-time parking app
youPark · React Native Developer
A React Native app for first-come, first-served parking: nearby free spaces on a map, distance, estimated arrival and how many other drivers are heading for the same spot.
Let's build something.
Have an AI feature, a product or a role to discuss? Book a 30-minute call, email me, or start a WhatsApp chat.