RemoteYeah
K
Kevin Yin
From United States 08:43 AM (GMT-05:00)
$160,000/yr
Active over a week ago
Member since May 2026

Share this profile:

Senior Machine Learning Engineer

Machine Learning Engineer
Available for hire
Years of experience
7+ years
Experience level
Senior
Available for
Full-time
Available from
07 Jul 2026

I'm a Senior Machine Learning Engineer with 10 years of production ML experience, currently building ads ranking models at Meta's Core ML team that power billions of daily impressions across Facebook, Instagram, Messenger, and WhatsApp. What makes me unique is the breadth of my journey through the ML lifecycle: I've shipped everything from ALS recommendation systems and time-series demand forecasts to today's transformer-based sequence learning models for pCTR/pCVR prediction. Before Meta, I spent years applying ML to entertainment and media at Vudu, Sony Pictures, and Amazon Prime Video—work that taught me how to connect modeling decisions directly to business outcomes, whether that was driving 50%+ of platform revenue through personalization or forecasting content demand for multi-million-dollar licensing deals. I'm looking for a senior or staff-level ML engineering role where I can own ranking and personalization systems end-to-end, from feature pipelines to real-time inference, and mentor others while shipping models that move real metrics at scale.

Skills

Python

Languages

No languages.

Employment History

Machine Learning Engineer – Ads Ranking, Core ML at Meta Current 2022 - Now
As an ML Engineer on Meta's Ads Ranking / Core ML team, I ship PyTorch ranking models powering billions of daily ad impressions across Facebook, Instagram, Messenger, and WhatsApp, including leading the transition from DLRM aggregations to transformer-based sequence models that delivered 2–4% incremental conversion lift. I built multi-task pCTR/pCVR prediction models, calibration pipelines (isotonic regression, Platt scaling), and Event-Based Feature pipelines over billions of daily events. I also run end-to-end A/B experiments and production monitoring via FBLearner Flow and Scuba to catch drift and ensure reliable, well-calibrated serving at scale.
Data Scientist at Vudu 2019 - 2021
As Staff Data Scientist at Vudu (Walmart/Fandango), I built a platform-wide personalized recommendation system using PySpark and Spark MLlib (ALS matrix factorization) that drove over 50% of revenue and streams while expanding coverage from 8 to 15 app platforms. I shipped personalized content row ranking (+12% CTR, 8% faster time-to-first-stream) and propensity models powering email recommendations (30% lift in sell-through), alongside real-time Kafka/Spark Streaming ETL pipelines and time-series demand forecasting. My personalization work helped lift the iOS app rating from 2.8 to 4.7 as the platform became fully personalized.
Staff Data Scientist at Walmart 2019 - 2020
As Staff Data Scientist at Walmart Global Tech, I built recommendation and ranking models using PySpark (Spark MLlib) and scikit-learn for large-scale personalization across Walmart's digital and streaming ecosystem, designing Spark-based preprocessing, feature generation, and batch inference pipelines over Hive tables on HDFS. I developed demand forecasting models (statsmodels, scikit-learn) producing weekly and monthly forecasts for content acquisition and inventory planning, orchestrated via Airflow, and built Kafka-based real-time event ingestion pipelines with Spark Streaming to enable near-real-time feature refresh for recommendation serving.
Data Scientist at Sony Pictures Entertainment 2016 - 2019
As a Data Scientist at Sony Pictures Entertainment, I built film green-light prediction models (scikit-learn, XGBoost, R) on box office, genre, cast, and release-calendar features to project revenue for film packages evaluated at $10M–$100M+ investment levels, with outputs reviewed weekly by finance and studio executives in Tableau. I also developed customer lifetime value models, recommendation engines (collaborative filtering, SVD), and audience segmentation models that guided marketing budget allocation and campaign targeting across millions of customers, plus theatrical and home entertainment revenue forecasting consumed by finance and distribution planning.

Education

Master's at University of California - Los Angeles 2011 - 2013
Report this profile
Browse more Machine Learning Engineer profiles