Builders want to go from intent to production-ready customer experience without leaving their AI assistant — but they work across multiple proprietary rendering frameworks that exist in no model's training data
Led vision, architecture, and hands-on delivery of an MCP-backed knowledge platform with per-framework grounding, embedded in IDE assistants — live framework guidance, reusable-component discovery, and automated code generation
Collapsed feature delivery from 3–4 weeks to 2–3 days for 200+ engineers
Architected an open agent registry where teams publish, fork, and share composable agentic workflows — so one team's codegen solution is discoverable and reusable by the next instead of rebuilt from scratch
Grew coverage from a single rendering framework to the full multi-framework stack, positioning the platform as the default AI authoring layer for the entire storefront engineering organization
Added validation so generated code meets compliance and accessibility standards before it ships
Led 2 engineering teams (15+ engineers), driving technical direction and org-wide adoption of AI-assisted development
Architected accessibility solutions using agentic AI patterns and
centralized knowledge management, supporting 100+ teams across Amazon's ecommerce
platform
Improved Accessibility Agent accuracy by 40% through advanced RAG architecture and centralized knowledge management
Built the first-ever accessibility metrics framework for Amazon Stores customers, measuring shopping effort across mobile and web
Led cross-functional collaboration with applied scientists, ML engineers, product managers, and stakeholders from concept to deployment
Led and mentored 15+ engineers across 3 teams in a GenAI adoption program
A documentation platform serving 40+ teams was suffering from its own success — slow manual onboarding, high publishing latency, no search integration, and growing operational cost
Concluded the right answer was not building something new but deprecating the platform entirely — after evaluating five alternatives and prototyping integrations to validate the assumptions
Made a deliberate “reuse over reinvent” call: 20 extra engineering days to migrate onto an existing hosting platform with 10+ aligned capabilities, consolidating two redundant products and eliminating ongoing dual-maintenance
Delivered 600% adoption growth and brought onboarding from days to under 30 minutes
Led a 4-engineer team to migrate 40+ documentation sites via one-click automation with zero downtime
Built a cross-build-system deployment tool along the way that became reusable by other teams
Negotiated an ownership model with the host platform team that remains the operating agreement today
Platform StrategyBuild vs. BuyMigration AutomationDeveloper OnboardingCross-Team Ownership
2019–2021
Amazon · Software Development Engineer II
People Engine & HCM Migration
Led development of critical HR systems and the process improvements
around them
Designed a unified solution measuring data staleness across People Engine services, supporting both legacy and new HCM systems
Implemented the foundational architecture for migrating 100+ employee-related integrations from Oracle HCM, reducing errors across badging and payroll
Built consensus across multiple teams for a streamlined solution while maintaining data accuracy