Machine Learning Engineer
AI Platform Corp.
Jun 2022 - Present
- Built recommendation engine serving 10M+ daily predictions with 99.5% availability and sub-80ms P99 latency
- Developed end-to-end ML pipeline from data ingestion to model deployment using Kubeflow and MLflow, reducing deployment time from 2 weeks to 3 hours
- Improved click-through rate by 25% through A/B tested model architecture changes across 4 product surfaces
- Designed and implemented feature store serving 500+ features to 12 production models, reducing feature engineering duplication by 70%
- Mentored 3 junior ML engineers and established model review process adopted across the ML platform team