Overview
Build and operate production machine learning systems for identity, audience intelligence, advertising, personalization, forecasting, engagement, and retention use cases across consumer platforms.
What you'll do
- Build low-latency serving systems for fraud scoring, message decisioning, and real-time personalization.
- Develop models for identity resolution, audience intelligence, content affinity, genre preference, and forecasting.
- Implement feature pipelines combining streaming signals with batch-computed features.
- Productionize models with experiment tracking, versioning, deployment automation, monitoring, and drift detection.
- Implement offline and online evaluations with baselines, metrics, and promotion criteria.
- Create design documents, readiness reviews, runbooks, and postmortem notes.
- Support junior engineers through guidance, code reviews, and knowledge sharing.
What you'll need
- 3–5 years of ML engineering experience, or 2+ years with a Ph.D.
- Experience owning production ML components across feature engineering, serving, monitoring, and iteration.
- Hands-on experience with Databricks, Spark, SageMaker, Python, and SQL.
- Experience deploying models for large user populations.
- Proficiency with PyTorch, TensorFlow, XGBoost or LightGBM, and scikit-learn.
- Bachelor’s or master’s degree in computer science, Statistics, Machine Learning, or a related field, or equivalent industry experience.
- Strong communication skills across technical and business audiences.
Nice to have
- Experience in identity resolution, audience modeling, recommendation or ranking, content understanding, streaming, fraud, or ad-tech ML.
- Experience with real-time feature serving, low-latency inference, mixture-of-experts, or graph neural networks.
- Published research or conference presentations in relevant ML domains.
Details
- Location: Hyderabad, India.
Read the full description and apply on the company’s own careers page.
NewPosted yesterday
