📍 Greater Seattle Area | ✉ sshreyv@gmail.com

Visit shreyshah.dev to explore my portfolio, ask my AI agent, and see my AI-native development work.

name: Shrey Shah
located_in: Greater Seattle Area
job: Senior AI SDE @ Microsoft | AI Agent Developer | Harness Engineer
public_voice: "I teach AI-assisted coding and agents"
ambassador_roles: ["Claude", "Cursor", "v0"]
education: ["Master of Engineering - University of Windsor",
            "Bachelor of Engineering - Gujarat Technological University"]
experience: "9+ years in Software Development | 3+ years in AI Engineering | 5000+ hours with AI coding tools"

current_roles:
  - ["Senior AI SDE", "AI Agents, Harness Engineering & Microsoft Copilot", "Microsoft", "Greater Seattle Area", "2026-Present"]
  - ["Claude Ambassador", "AI-Assisted Coding & Agent Workflows", "Anthropic", "Remote", "Present"]
  - ["Cursor Ambassador", "AI Coding Workshops & Community Feedback", "Cursor", "Remote", "Apr 2025-Present"]
  - ["v0 Ambassador", "Full-Stack AI Prototyping", "v0 by Vercel", "Remote", "Dec 2025-Present"]

past_experiences:
  - ["Senior AI Software Engineer", "Multi-Agent Systems, RAG & Memory Architectures", "Vivun", "Canada/US", "May 2023-Mar 2026"]
  - ["Lead SDET / Test Architect", "Test Automation & Backend Development", "Priceline", "USA", "June 2021-May 2023"]
  - ["Automation Developer", "Test Framework Development", "Powerhub Inc", "Canada", "Sept 2020-June 2021"]
  - ["QA Automation Engineer", "Blockchain, Mobile & IoT Testing", "NXM Labs", "Canada", "Feb 2019-Apr 2020"]

fields_of_interest: ["AI Agents", "Harness Engineering", "LLM Evals",
                     "Context Engineering", "Multi-Agent Workflows", "RAG Systems",
                     "Knowledge Graph Memory", "MCP Development", "AI-Assisted Coding",
                     "Self-Healing Systems", "Voice Agents"]
technical_background: ["LangGraph", "LangChain", "FastAPI", "Next.js", "TypeScript",
                       "Python", "Microsoft Copilot", "Claude Code", "Cursor", "v0",
                       "DSPy", "Vector Databases", "Knowledge Graphs", "ComfyUI",
                       "MCP Servers", "Test Automation", "CI/CD Pipelines"]
currently_learning: ["Agent Harness Design", "Production Evals", "Computer Use Agents", "AI Safety"]
hobbies: ["VR Gaming", "Photography", "AI Art", "Marvel & DC", "Tech Gadgets"]
achievements: ["Senior AI SDE @ Microsoft", "Claude/Cursor/v0 Ambassador",
               "Hosted AI coding workshops for thousands of developers",
               "Built full-stack apps live with AI-assisted workflows",
               "Created MCPs and agent tooling for dynamic development",
               "Open-sourced AI coding resources and personal agent workflows"]

Architecting multi-agent systems with LangGraph and production guardrails Building agent harnesses for repeatable planning, execution, review, and eval loops Designing RAG + memory systems with knowledge graphs and long-term context Creating MCP servers and agent-native tools for real workflows Optimizing prompts and workflows with evaluation-driven development Testing agent systems for prompt injection, reliability, and failure recovery

5000+ hours with AI coding tools across Copilot, Cursor, Claude Code, and v0 Teaching developers how to build with agents instead of one-off prompts Building full-stack apps live with AI-assisted workflows Turning tickets, docs, and voice notes into PR-ready implementation loops Creating reusable Cursor rules, Claude workflows, and agent playbooks Helping teams adopt AI tools with guardrails, reviews, and evals

📦 Projects	⭐ Stars	🔀 Forks	💻 Language
Awesome Claude Code Mastery			Resources
ComfyUI Themes			TypeScript
Awesome AI Agents			Python
LangGraph Interview Questions			Resources
Cursor Hub			Shell
unclaw			Shell

"Building the AI-native future: agents that learn, self-improve, and make software teams faster without sacrificing reliability."

I'm a Senior AI SDE at Microsoft, an AI Agent Developer, and a Claude, Cursor, and v0 Ambassador. I talk about harness engineering, AI-assisted coding, and the shift from prompt demos to reliable agent systems.

My background in test architecture shaped how I build AI: every workflow needs guardrails, evals, feedback loops, and failure recovery. Today I focus on building agent systems that can plan, execute, review, and improve with humans in the loop.

• 🤖 Senior AI SDE @ Microsoft working on AI agents and Microsoft Copilot-oriented workflows
• 🧭 Claude, Cursor, and v0 Ambassador teaching AI-assisted development and agent workflows
• ⚡ 5000+ hours building with AI coding tools from early GitHub Copilot to modern agentic IDEs
• 🎤 Hosted workshops, conference sessions, and live demos for thousands of developers
• 🔧 Created MCP servers, Claude/Cursor workflows, and agent-native tooling for dynamic development
• 🧪 Advocating for evaluation-driven development so AI systems can be trusted in production

• 🧰 Designing agent harnesses for planning, tool use, review, and self-healing loops
• 🧠 Building memory systems with knowledge graphs, retrieval, and long-term context
• 📖 Developing advanced RAG systems: hybrid RAG, GraphRAG, and agentic RAG
• ⚗️ Optimizing agent quality with LLM-as-a-judge, evals, and adversarial testing
• 🔌 Creating MCP servers and workflow automation for real developer environments
• 🛡️ Making AI-assisted coding safer with guardrails, review agents, and reproducible workflows

When I'm not building AI systems, you'll find me:

• 🎮 Exploring virtual worlds in VR
• 📸 Capturing moments through photography
• 🎨 Creating AI art with Stable Diffusion and ComfyUI
• 🎬 Watching Marvel and DC stories
• 📱 Testing the latest productivity and AI gadgets

---