ChipotleReimagined

Project Concept

No description has been added yet.

Entry

Status: Submitted

Last saved: November 15 at 6:31 PM EST

View Entry

Team Roster

Message board not available for this team yet.

Hubert Leo Team Lead RSVP Approved

Analyst at NYU Langone

(Full-stack & Frontend) Chrome Extension architecture (MV3): content scripts, service worker, DOM injection; persistent state via Chrome Storage API. Menu transformation engine: dynamic UI swap with MutationObserver (handles Chipotle’s live DOM updates) and section visibility via IntersectionObserver. Multi-cuisine system (6 profiles): config-driven items (names, images, descriptions), hero replacements, dropdown switching. Food card mapper: maps data-analytics-section → cuisine items; memory-safe tracking with WeakMap. Asset pipeline: image replacement via chrome.runtime.getURL(); CSS variables for consistent theming. Voice system scaffold: namespacing (window.VoiceSystem), section detection, click delegation, and selection state machine (protein → rice → beans → toppings). Backend foundation: Express.js server with CORS, error handling, and graceful Redis fallback. Tech: Chrome Extension APIs (MV3), JavaScript (ES6+), DOM APIs, MutationObserver, IntersectionObserver, Express.js, Chrome Storage API, CSS3, HTML5.

Researcher at NYU Langone.

Fascinated by AI agents/productivity tools in your desktop since watching a demo of Computer Use. Currently reading up on literature on GUI agents and exploring open source tools for Computer Use.

Working on AI productivity tools. Right now working on a meeting companion tool that researches highlighted regions on the screen automatically.

William Tjandra RSVP Approved

Student at Ucla

(Backend & Voice) Tavily integration (sponsor API): backend Express route /api/food-info issuing parallel queries (general + cultural); response shaping for UI. Redis caching (sponsor tech): redis client with connection pooling; 30-day TTL; normalized cache keys per cuisine (food-italian-beef-stracotto) → 90%+ Tavily call reduction. Info UX: item info button injected by content script; modal renders Tavily summaries + source links; messaging via chrome.runtime.sendMessage to background service worker. Background worker: proxy fetch with AbortController (15s timeout), error propagation, and retry policy. Voice ordering logic: first-scroll trigger, one-time prompt guard (hasPrompted), sequential prompts (protein → rice → beans → toppings), and skip detection (fallback if rice/beans/toppings clicked before protein). Audio pipeline: ElevenLabs API with automatic fallback to local MP3s; blob URL lifecycle; autoplay policy handling (gate on user interaction). File org: cuisine-scoped voice assets (e.g., italian/choose-protein.mp3, fallback/skipped-protein.mp3) mapped in playVoice.js. Tech: Redis, Tavily API, Express.js/Node.js, ElevenLabs API, HTML5 Audio API, Chrome Background Service Workers, Chrome Runtime Messaging, IntersectionObserver, AbortController.

I study Mechanical Engineering at UCLA, but I love making projects on the side and building "startups". My dream is to start and run a successful startup.

Ai/ml