John Peng

Senior Data Scientist at C3 AI
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
You need to have a working account to view this content.

Experience

    • United States
    • Software Development
    • 100 - 200 Employee
    • Senior Data Scientist
      • Jun 2022 - Present

    • Data Scientist
      • Jan 2021 - May 2022

      Predictive Maintenance• Designed and wrote pipeline using Python and the C3.ai platform to conduct machine learning experiments remotely and asynchronously. Process included retrieving data, transforming data, splitting data into train/validation/test sets, automatically tuning hyperparameters, and persisting predictions.• Predicted likelihood of subsystem failure using operational history and sensor readings

    • Associate Data Scientist
      • Oct 2019 - Jan 2021

      Energy Management• Trained and tuned boosted tree models to predict chilled water and steam usage at a university campus• Predicted multiple days with the highest electricity usage throughout a power grid to help customer reduce concurrent peak charges

    • United States
    • Strategic Management Services
    • 1 - 100 Employee
    • Data Scientist and Information Manager
      • May 2019 - Sep 2019

    • United States
    • Higher Education
    • 200 - 300 Employee
    • Data Science
      • Jan 2019 - Apr 2019

      Capstone Project – Predicting fire risk from building modifications github.com/porbo/sf-fires • End-to-end project, from data acquisition and cleaning to delivering predictions and presenting • Predicted fire risk using logistic regression on tfidf vectorization of building modification descriptions • Predicted indirectly from unavailable data like building age or population density • Geocoded addresses to visualize predictions with plotly Case Studies • Detected over 95% of fraudulent event postings from an event management website, with sub-20% false positive rate. • Implemented collaborative filtering to recommend movies receiving an average rating of 4.27/5 • Predicted tractor prices using ridge regression with cross-validated hyperparameters • Predicted churn in ride-sharing using gradient-boosted decision trees, and recommended targeted action based on churn-causing factors Show less

    • United States
    • Software Development
    • 700 & Above Employee
    • Tutor
      • Sep 2016 - Dec 2018

      • Worked with students/clients to prepare custom lessons based on their needs ◦ Received 5 star rating as a tutor ◦ Devised reading drills which noticeably improved practice test performance • Tutored students in: high school algebra, SAT preparation, AP Statistics • Worked with students/clients to prepare custom lessons based on their needs ◦ Received 5 star rating as a tutor ◦ Devised reading drills which noticeably improved practice test performance • Tutored students in: high school algebra, SAT preparation, AP Statistics

    • China
    • Technology, Information and Internet
    • 700 & Above Employee
    • Intern, E-Commerce BI
      • Jul 2014 - Sep 2014

      • Evaluated e-commerce websites for strategies, value, and partnership opportunities ◦ Recommended investing in dealmoon.com (has continued to grow) over deal.com.sg (stopped operating) ◦ Researched companies and websites include Dealmoon, Deal.com.sg, Gome, dealnews, Fancy, JingDong (JD) • Conducted market research on international e-commerce ◦ Wrote a comprehensive overview of cross-border e-commerce to lay groundwork for potential new lines of business Show less

Education

  • University of California, Los Angeles
    Actuarial Science
    2013 - 2017
  • University of California, Berkeley
    N/A, N/A
    2018 - 2018

Community

You need to have a working account to view this content. Click here to join now