Staff Machine Learning Engineer - (Growth & ML Team), Hyderabad

You will set the technical direction for the team’s most complex and strategically important ML systems, serve as the technical authority across multiple concurrent workstreams, and act as a force multiplier for the entire team. This role requires 8+ years of experience, exceptional depth across the ML stack, and the ability to influence technical decisions across organizational boundaries. Define and own the technical architecture for the team’s core systems: the probabilistic identity spine, audience intelligence platform, content-affinity and genre-preference models, and ML-based forecasting. Lead architectural decisions for the team’s MLOps framework — feature-store design, training-pipeline standards, model-serving patterns, and monitoring infrastructure — on a Databricks-first architecture, integrating Snowflake and AWS SageMaker where each is the right tool. Evaluate and recommend new technologies and approaches (e.g., DCR-native modeling, graph ML, agentic ML orchestration, LLM-augmented pipelines) with clear build/buy/partner assessments. Drive standardization of ML practices across Hyderabad and align with global WBD ML engineering standards. Architect and lead delivery of the probabilistic identity resolution system — resolving unauthenticated device IDs and 1P cookies to households/persons with calibrated confidence at scale across all WBD brands — using entity resolution, embeddings/representation learning, calibration, candidate blocking, and champion/challenger promotion, with a roadmap toward person-level graphs (incl. GNNs) and real-time resolution. Lead the evolution of Audience Intelligence: ML Promo Optimizer (layered retrieval + propensity + closed-loop measurement, toward mixture-of-experts), STAT v2 (two-tower single-title affinity with semantic content embeddings), lookalike modeling (LAL 2.0+) inside Snowflake DCR, and content segmentation. Own the ML architecture for forecasting (audience growth, demand, yield/pricing) and ensure models are production-grade, monitored, and continuously improved. Drive the roadmap for bringing ML personalization signals (genre/content affinity, engagement trends) into batch and, over time, real-time activation paths. MLOps Platform Leadership. Define the team’s MLOps target architecture: feature contracts, model-registry governance, automated retraining, drift detection, and A/B experimentation infrastructure. Establish engineering standards for the full ML lifecycle: data contracts → feature engineering → training → evaluation → deployment → monitoring → deprecation. Champion a lightweight, adoptable, well-documented architecture (Databricks Asset Bundles, GitHub Actions CI/CD, MLflow throughout, feature tables, inference/monitoring tables), with leakage prevention, reproducibility, and FinOps controls baked in. Serve as the technical anchor for Hyderabad; provide architectural guidance and deep mentorship to Senior and MLE 2 engineers. Partner with the Senior ML Engineering Manager on technical roadmap prioritization, headcount planning, and team capability development. Represent the Hyderabad team in cross-functional technical forums with US-based ML, Data Engineering, Product, and Ad Sales stakeholders. Produce high-quality technical design documents, architecture reviews, and post-mortems that elevate the team’s engineering culture. 8+ years of industry experience in ML engineering (6+ with a Ph.D.), with demonstrated Staff-level scope and impact. Mastery of the full ML stack: data engineering, feature engineering, model development, MLOps, and production monitoring. Deep Databricks expertise: Delta Lake, Unity Catalog, Workflows/DLT, MLflow, Feature Store, Asset Bundles, and Genie Space configuration. Strong AWS proficiency (SageMaker training/pipelines/model registry, S3, Lambda, Glue) and Snowflake expertise (DCR patterns, Snowpark, Cortex, SQL optimization at scale). Proven experience architecting production ML systems serving millions of users, and a track record of technical leadership (setting standards, driving architecture, influencing across teams). Expert proficiency with ML frameworks (PyTorch, TensorFlow, XGBoost/LightGBM, scikit-learn) and deep understanding of statistics and ML fundamentals. Master’s or Ph.D. in Computer Science, Statistics, Machine Learning, or a related field (or equivalent industry experience). Excellent communication, with the ability to advocate technical solutions to engineering, science, product, and executive audiences. Streaming / media / ad-tech ML: identity resolution, audience modeling, recommendation/ranking, content understanding. Deep knowledge of probabilistic identity resolution (entity resolution, graph-based household matching, confidence calibration) and Data Clean Room ML (Snowflake DCR, AWS Clean Rooms). Hands-on experience with agentic AI frameworks at production scale (LangGraph, AutoGen, MCP, CrewAI), Databricks Genie Space curation, and Snowflake Cortex Search / Fine-Tuning for enterprise RAG. Experience with real-time feature serving and low-latency inference, and with mixture-of-experts or graph neural networks. Primary platform: Databricks (Lakehouse, PySpark, Delta, Workflows/DLT, MLflow, Feature Store, Unity Catalog, Asset Bundles, Genie). Cloud: AWS (SageMaker, S3, Lambda, Glue). Warehouse: Snowflake (DCR, Snowpark, Cortex). Agentic AI: Cursor, GitHub Copilot, Amazon Q, Databricks Genie, Snowflake Cortex, MCP. Languages: Python (primary), SQL, Scala (as needed). We build primarily on Databricks, with strong working knowledge of Snowflake and AWS, and we are an early, enthusiastic adopter of agentic AI development workflows. What You’ll Do. Preferred. Championing Inclusion at WBD.