About Said Project
Said built a lightweight tool that transforms ordinary slide decks into delivery-ready prompts and micro-scripts aligned to MEET values. The tool is designed to cut nightly prep, keep co-teaching and binational collaboration visible in the moment, and standardize how instructors embed values into every slide not just a single “values” slide.
How did the program impact your perspective and understanding of machine learning?
These skills will with me whenever I want to build another project. For example I'm doing an internship an AI internship in Germany next month. This will be the base for me if I really make this um my goal and I put effort into remembering everything and memorizing and like understanding the machine learning that I learned in here.
From Concepts to Relationships
Rather than building a generic text rewriter, Said mapped relationships among three inputs: slide text, an instructor’s contextual notes, and the MEET value anchors. The tool generates five micro-output types per slide — hooks, punchlines, short acts, clarifying questions, and vibe-resets, each explicitly annotated with which MEET value it serves. This reframing moved the product from “automated copy” to “teaching scaffolds”: little, interpretable artifacts instructors can choose or adapt in-class.
“I wanted something that feels like a co-teacher, not a replacement, bite-sized prompts that respect the instructor’s voice and the class’s rhythm.”
Learning to Trust (and Question) Outputs
Early trials with two instructors exposed important trade-offs. Some generated hooks were too literal for certain cultural inflections; others missed subtle cues that a live instructor would notice.
Said added quick visual checks and a two-click edit flow so instructors can quickly swap tone or re-align a generated prompt with the slide’s intent. Adoption hinged less on perfect automation and more on predictable, editable outputs instructors could trust and reuse nightly.
Thinking in Systems, Not Screens
Said didn’t stop at a single-page generator. He designed an onboarding microflow for new instructors: a 10-minute orientation that shows how a slide’s prompt maps to a MEET value, a one-click export to slide notes, and a lightweight shared log that records which instructor used which prompt (preserving binational collaboration in the delivery record).
This system makes values auditable and reproducible, and it shortens onboarding time for external or substitute instructors.
Looking Back
The project is a pragmatic win: it shifts time from prep to presence. Proof-of-concept usage across ≈120+ students received positive feedback from two instructors and opened the door for a pilot next year. Expected outcomes include more engaged sessions with values embedded in hooks and acts, faster instructor onboarding, and measurable minutes reclaimed for mentorship and project work. Said plans to refine the edit controls and expand the MEET-value mapping to better support cross-cultural phrasing.
Ready to Apply?
Jump in and start your journey!
We encourage students to fill out the application themselves it gives us a clearer sense of their interests and intent. Please take a moment to read through the questions and answer them with care. Each application is reviewed thoughtfully, so genuine, well-considered responses really do make a difference.
