Projects / PennApps XXVI
Award winnerMLH - Best use of MongoDB AtlasConfidence: low
MovieMind
Recommends films for the reasons you like a movie, by matching your review to others' reviews.
Project facts
- Team
- 2 people
- Build
- software only
- AI
- llm api, ml model
Problem
Recommendation algorithms optimise watch time, not why people enjoy films.
Solution
Embeds user reviews, finds semantically similar reviews in a vector database, takes the most positive matches as recommendations, and has an LLM explain the reasoning.
Demo moment
Team-described: write why you loved a movie and get explained recommendations.
Why it stood out
- organizer statedWon “MLH - Best use of MongoDB Atlas” among 85 submissions.Source ↗
- team statedUses the sponsor's vector search as the core of the product.Source ↗
Lessons for your next hackathon
- Explaining a recommendation is a cheap way to make it feel human.
Similar situations
Projects that share domains, mechanisms or judging lenses — often with a different problem and stack.Tune the comparison.
Award winner · 2026
A space shooter that studies how you play and adapts enemies, a boss fight and help to you.
Shared: domains games entertainment · mechanisms sponsor platform mastery · lenses sponsor api use · ai llm api, ml model · stack react
Award winner · 2025
A companion that answers viewers' questions about a movie in real time without spoilers or interruptions.
Shared: domains games entertainment · lenses sponsor api use · ai llm api · stack react, gemini
Award winner · 2025
Agentic AI characters inside the game Terraria, spawning many agents at once.
Shared: domains games entertainment · lenses sponsor api use
Award winner · 2025
A risk-free stock-market simulation game where young investors race to beat the market.
Shared: domains games entertainment · mechanisms sponsor platform mastery · lenses sponsor api use · ai llm api · stack react, cerebras
Award winner · 2025
Answers a new engineer's 'why' questions with sources from Slack, GitHub, Linear and Notion.
Shared: mechanisms cited evidence trust, sponsor platform mastery · lenses sponsor api use · ai llm api · stack next.js, react, gemini
Award winner · 2025
A real-time multiplayer tank battle where players stake a stablecoin in escrow and the ledger pays the winner.
Shared: domains games entertainment · mechanisms sponsor platform mastery · lenses sponsor api use · stack node.js, mongodb
Sources
- https://devpost.com/software/moviemind-85pfcw · devpost project · retrieved 2026-10-13
- https://pennapps-xxvi.devpost.com/project-gallery · devpost gallery · retrieved 2026-10-13