Skip to content
John Quevedo

John Quevedo

CS + Math student at Yale interested in machine learning and algorithms.

Resume PDFGitHubLinkedInEmail
John Quevedo hiking on a mountain trail.

Featured Projects

View all
Probabilistic stock volatility forecasting dashboard with prediction intervals and trend chart.
MLSystemsResearch

Probabilistic Stock Volatility Forecasting

Built a machine learning pipeline to forecast 5-day stock volatility using historical market data.

  • - Improved forecast accuracy by 23–28% compared to simple baseline methods across SPY, QQQ, and AAPL.
  • - Shipped a Streamlit dashboard for visualizing forecasts, uncertainty bands, and model performance.
#Python#PyTorch#GPyTorch#Streamlit#yfinance
Case studyRepoDemo
MovieMatch Streamlit interface with search, ratings, and recommendations panels.
MLResearch

Movie Recommendation System

Built a personalized movie recommendation system using collaborative filtering on the MovieLens 100K dataset.

  • - Increased catalog coverage from 3.9% to 37.0% of movies.
  • - Deployed an interactive Streamlit app that updates recommendations in real time from user-ratings.
#Python#implicit#Pandas#NumPy#SciPy#Streamlit
Case studyRepoDemo
Shelf landing page showing social reading features and interface preview.
SWESystems

Full-Stack Social Reading Platform

Built a full-stack social reading app with search, personal bookshelves, reviews, follows, and activity tracking.

  • - Designed PostgreSQL/Prisma schema and APIs for authentication, notifications, comments, likes, and follows.
  • - Implemented responsive UI and secure authentication, enabling persistent user profiles and social interactions.
#Next.js#TypeScript#Tailwind#PostgreSQL#Prisma#NextAuth#Docker
Case studyRepoDemo

Experience

Yale University — New Haven, CT

Undergraduate Researcher, Transformer Reasoning — Feb 2026 – Present

  • - Ran controlled training experiments to analyze emergence of in-context learning and chain-of-thought reasoning.
  • - Built evaluation pipelines to test effects of data distribution and curriculum schedules on reasoning.
  • - Implemented looped transformer variants and measured changes in iterative reasoning behavior.

Yale University — New Haven, CT

Undergraduate Researcher, Parallel Graph Algorithms — Oct 2024 – Present

  • - Built parallel batch-dynamic k-clique counting for large sparse graphs on multicore systems.
  • - Optimized parallel hash tables and set intersections to reduce update latency and improve scaling.
  • - Cut memory by replacing quadratic intermediates with arboricity-aware sparse representations.

Skills & Tech

TypeScriptPythonNext.jsReactTailwindPostgreSQLPrismaPyTorchGPyTorchDocker