Meera Raghunathan
Machine Learning Engineer · Ranking & Recommendations
Professional Summary
Machine learning engineer with 4 years building ranking and recommendation systems for e-commerce. Takes models from offline experiments to low-latency production serving, with monitoring and retraining in place. Lifted add-to-cart rate by 6.8% with a two-tower retrieval model.
Work Experience
Machine Learning Engineer
Saffronleaf Commerce · Bengaluru
- Built a two-tower retrieval model in PyTorch over 12M products, raising add-to-cart rate by 6.8% in an A/B test across 4M users.
- Cut feature-store lookup time from 38ms to 9ms by moving hot features to Redis and batching requests.
- Set up drift monitoring and weekly retraining with Airflow and MLflow, catching 3 data-pipeline breaks before they reached users.
- Served models on Kubernetes with Triton at 2,500 requests/second and p99 latency under 60ms.
Data Engineer
Tidewell Logistics · Chennai
- Built Spark pipelines that cleaned 40M shipment events a day for the delivery-time model.
- Wrote the first gradient-boosted ETA model, reducing mean absolute error from 41 to 27 minutes.
Education
B.Tech, Computer Science and Engineering
Adyar Estuary Institute of Technology · Chennai · Score: CGPA 8.7/10
Skills
ML & Deep Learning: PyTorch·TensorFlow·scikit-learn·XGBoost·Hugging Face
Data & MLOps: Python·SQL·Spark·Airflow·MLflow·Feast
Serving & Cloud: Docker·Kubernetes·Triton·Redis·AWS SageMaker
A/B testing·Experiment design·Mentoring
Download your resume to print it.
Aria
ATS-safeSingle-column, plenty of whitespace, parses cleanly everywhere. A safe default.
- No photo — text only.
- Page size: A4.
- Switching to another template later keeps all your content.