Curriculum data was scattered across hundreds of documents, which made it hard to search, compare, or analyze in any meaningful way. I designed and built BOLT to turn that mess into a practical system teams could actually use.

The Problem

Important curriculum information lived in separate files, across departments, with inconsistent structure and no simple way to surface patterns or answer questions quickly. Finding where something was taught, identifying overlap, or spotting gaps required too much manual digging.

The real problem was not just storage. It was usability. The information existed, but it was buried.

The Goal

Build a system that could pull curriculum data into one place, make it searchable, and support smarter analysis without forcing users to learn a complicated technical workflow.

• Create a usable search experience across many curriculum documents
• Structure messy source data into something consistent
• Use AI where it added value, not just for show
• Keep the output clear, fast, and useful for real educators

My Role

I led the system design, defined the workflow, shaped the user experience, and used AI to help fill technical gaps while building. That included planning the data structure, refining the search behavior, testing output quality, and iterating on how results were presented so they felt helpful instead of robotic.

The Approach

I broke the project into practical layers instead of trying to solve everything at once.

• Parsed curriculum maps into structured data
• Built a searchable system that could return useful matches
• Used AI-assisted workflows to accelerate development and problem-solving
• Improved formatting and summaries so the output was readable and grounded
• Continually tested whether the system answered real questions in a way people could trust

What I Built

BOLT became more than a search tool. It evolved into an AI-assisted curriculum management system that could help surface where topics appeared, support curriculum review, and make existing documents more accessible to a larger team.

Instead of asking users to dig through folders and guess which files mattered, the system was designed to bring relevant information forward in a structured, teacher-friendly way.

Outcome

The result was a working system that made curriculum information more visible, more searchable, and far more usable. It reduced the friction of finding answers buried across many documents and created a stronger foundation for future analysis, including identifying gaps, overlaps, and patterns across courses.

Why It Matters

This project reflects the kind of work I do best: turning a messy, high-friction problem into a practical system people can actually use. It also shows how I work with AI in a grounded way, not as a gimmick, but as a tool to move faster, think better, and build useful things.

Key Takeaways

• I build systems around real use cases, not abstract features
• I use AI to extend my capabilities and speed up iteration
• I care as much about clarity and usability as I do about functionality
• I am strongest when translating complexity into something practical