Class Projects

During my Management Information Systems degree at Baylor, two courses in particular pushed me from theory into genuine applied work. Professor Bhojwani's SQL and AI Implementation class gave me a structured introduction to database design and query optimization alongside hands-on exposure to generative AI tooling. The projects in that class were the first time I built something real with LLMs rather than just reading about them.

The centerpiece project was a LangChain-powered research assistant built on top of a Django backend, designed to answer domain-specific questions by retrieving context from a structured SQL database and passing it to an LLM for synthesis. The system used a retrieval-augmented approach where user queries were parsed, matched against indexed records, and returned as grounded natural-language responses rather than hallucinated outputs. A secondary project used n8n to automate a multi-step data ingestion and classification workflow, connecting an external API, a local database, and a notification layer without writing custom integration code for each step.

These projects collectively shifted how I think about building with AI - less as a magic layer you drop in, and more as a structured component in a larger data pipeline. They also gave me a working vocabulary for the tools I now reach for on real client work, including Shepherd and Argus.