J

Jason Ting

About

Detail

Louisiana, United States

Timeline

work
Job
school
Education

Résumé

Jobs verified_user 0% verified
  • Plaid
    Machine Learning Engineer
    Plaid
    Nov 2021 - Nov 2025 (4 years 1 month)
    Engineered and deployed core machine learning models powering the Plaid Protect and Plaid Signal platforms, essential for real-time fraud detection, identity verification, and transaction risk scoring within the global FinTech ecosystem. Developed and iterated on advanced proprietary ML models, including the Trust Index (Ti1 and later Ti2), leveraging thousands of signals, network-wide graph analysis, and transaction history patterns using Python 3.9+. Utilized core ML frameworks (PyTorch and TensorFlow 2.x) for model training and deployed models into a highly scalable, event-driven architecture built primarily on Go and Python microservices. Managed the entire MLOps lifecycle on a robust AWS infrastructure, orchestrating services across te
  • TikTok
    Machine Learning Engineer
    TikTok
    Sep 2020 - Oct 2021 (1 year 2 months)
    Contributed to the development and optimization of TikTok's globally renowned, hyper-personalized recommendation algorithm, which serves content to hundreds of millions of users in real-time. Engineered and fine-tuned large-scale AI models for core functionalities, including content understanding (NLP and Computer Vision), user behavior modeling, and search ranking. Utilized core deep learning frameworks PyTorch and TensorFlow 2.x for training multi-modal and multi-head models on massive datasets, running A/B tests on proprietary experimental platforms. Developed robust data pipelines using a hybrid batch and stream processing architecture, leveraging Apache Kafka for real-time data streaming and Apache Spark/Flink for processing massive vo
  • Airbnb
    Machine Learning Engineer
    Airbnb
    Aug 2019 - Jul 2020 (1 year)
    Developed and deployed machine learning models that personalized user experiences across key platform areas, including search ranking, pricing recommendations (Smart Pricing), and host/guest matching. Leveraged Airbnb's internal end-to-end ML platform, Bighead, to manage the full model lifecycle, utilizing Python as the primary development language and executing jobs on internal infrastructure powered by TensorFlow and PyTorch. Engineered robust data pipelines for feature engineering and data preparation, utilizing Apache Spark and Presto (data querying engine) for distributed processing of massive datasets within the Hadoop/Hive ecosystem. Managed data storage within an internal data lake powered by HDFS and performed analysis using SQL (v
  • Facebook
    Software Engineer, Data Scientist
    Facebook
    Jun 2015 - Jul 2019 (4 years 2 months)
    Developed and deployed sophisticated machine learning models that drove core platform functionalities, including News Feed ranking, ad targeting optimization, and user engagement metrics for hundreds of millions of daily active users (DAUs). Utilized Python and R for rapid experimentation and analysis, implementing models using internal, large-scale ML platforms powered by deep learning frameworks like PyTorch and early versions of TensorFlow/Caffe2. Engineered massive-scale data pipelines using internal infrastructure built upon the Hadoop ecosystem (HDFS, Hive, HBase), processing petabytes of user data using Apache Spark and Presto (data querying engine). Managed the entire ML model lifecycle using Facebook's proprietary FBLearner Flow pl
  • Quorum Analytics
    Data Scientist
    Quorum Analytics
    Jun 2014 - Sep 2014 (4 months)
    Developed data analysis pipelines to extract, clean, and analyze legislative and public policy data, providing actionable insights for clients in a civic tech environment. Utilized Python 2.7+ as the primary scripting language for data manipulation, leveraging libraries like Pandas and NumPy for statistical analysis and data wrangling within a standardized Linux Environment (likely Ubuntu/CentOS). Applied foundational machine learning algorithms (e.g., logistic regression for classification, k-means for clustering) using the Scikit-learn library to categorize and structure unstructured text data from legislative documents. Used SQL to query data stored in relational databases (PostgreSQL or MySQL) and employed robust web scraping techniques
Education verified_user 0% verified
  • Stanford University
    Master's degree, Computational Mathematical Engineer
    Stanford University
    Jan 2014 - Jan 2016 (2 years 1 month)
  • Harvard University
    Bachelor of Science, Mechanical Engineering, Computer Science
    Harvard University
    Jan 2010 - Jan 2014 (4 years 1 month)