Walker Stevens Email & Phone Number

Los Angeles, Ca, Us

• Used data from over 500,000 nationwide apartment listings to develop astacking model predicting rents with 8% MSPE, improving market predictionaccuracy by 50%.• Presented the model at the American Real Estate Conference, leading toacademic collaborations and enhancing the company's industry reputation.• Implemented a vision transformer in PyTorch using images to assess buildingquality, reducing manual evaluation time by 60%.• Developed and maintained automated pipelines using Databricks and MLflowfor continuous model training and deployment, boosting operational efficiencyby streamlining workflow processes.• Created statistical models to forecast office construction and leasing rates,providing actionable insights that informed millions of dollars in investmentdecisions.

• Collaborated with NASA IMPACT to enable searching petabytes of unlabeledimagery by building an embedding representation of Earth science data fromNASA Worldview and astrophysics data from Hubble Space Telescope.• Trained a self-supervised model, deployed on a production cluster, currentlyused by NASA scientists, achieving Technology Readiness Level 7.• Presented findings to NASA leadership, securing a top 5 position in the NASAScience Mission Directorate's grant competition, demonstrating strongcommunication and presentation skills.

• Automated data extraction and analysis for international physicsexperiments, such as the Simons Array, improving data processing speed by30%.• Improved hardware performance on telescopes by detecting and analyzingelectronic component changes, leading to more accurate observations.• Developed regression models for supernovae observations, contributing topublished research in Physical Review D.

Master Of Science - Ms, Analytics

Bachelor Of Science - Bs, Physics

I wrote several programs tracking the proper motion of asteroids in reference to the background of stars. I also wrote a program reducing.

Certificate, Data Science

A 13 week immersive with 700+ hours of coding, weekly Case Studies, and 3 capstones. Python-based curriculum focused on machine learning.