AI-Powered E-Commerce Automation Platform

Senior capstone project built in collaboration with an industry partner, centered on automating the migration of e-commerce storefronts between platforms using AI-assisted transformation pipelines. The system targets a real pain point: storefront migrations that typically require weeks of manual developer work, translating themes, configurations, and product data from one platform's conventions to another. My primary responsibility was designing and implementing the execution and validation layer — a sandboxed runtime that safely processes dynamically generated transformation scripts, enforces output contracts, and deterministically handles partial failures. This required careful attention to security boundaries (preventing untrusted code from escaping the sandbox), repeatability (same input always produces the same output), and auditability (structured logs at each pipeline stage). The system was built with C# and .NET, integrating with cloud services for storage and orchestration. Specific internal architecture details are NDA-constrained, but the architecture poster and demo video illustrate the overall pipeline design.