Amazon
Software Development Engineer Intern
May 2026 - Aug 2026 · Toronto, ON
At Amazon, I helped create a route-planning optimization algorithm now deployed in every Amazon fulfillment center worldwide. The demo below picks the same items along a shorter route.
A shorter way through the warehouse
Illustrative demoOriginal route · 58 grid steps · 0/3 items picked
Illustrative routes: 58 to 46 grid steps. Not Amazon’s production system or a performance benchmark.
What I worked on
Decomposed a 9,400-line warehouse decision engine into pluggable microservices, enabling dynamic selection of item-picking optimization algorithms deployed across every fulfillment center.
Reduced legacy service migration time by 98% (two weeks to two hours) by building an agentic AI migration system with autonomous architecture discovery and 20 reusable deterministic transformation skills.
Re-architected the execution engine across ECS, Lambda, Step Functions, and S3, eliminating redundant computation in algorithm selection and reducing end-to-end execution time from 16 seconds to 4.5 seconds.
Built a semantic-diff validation harness to replay production JSON payloads against generated endpoints, evaluating behavioral parity via a confidence metric that flags inconclusive noise across 42 execution paths.