Machine Learning Engineer – Recommendation Systems / Propensity Modeling

Role Overview

We are looking for a Machine Learning Engineer / Data Scientist to design and build a high-scale offer recommendation system that personalizes and ranks offers for millions of users to improve engagement and conversion. The role focuses on propensity modeling, ranking systems, and personalization in a dynamic user environment with limited and noisy data.

Business Context

·       Platform displays multiple third-party offers (10–20 offers per user)

·       Goal: Show top 1–3 offers and maximize CTR, conversion, and engagement

·       Scale: Millions of users with dynamic and short-lived user base

·       Constraints: Limited user history, cold start, privacy constraints (no third-party data), sparse data

Key Responsibilities

Recommendation System Design:

·       Design end-to-end offer recommendation pipeline

·       Build personalized ranking systems for user-offer matching

Propensity Modeling & Ranking:

·       Predict CTR and engagement probability

·       Use models like XGBoost / LightGBM

·       Optimize ranking using NDCG, MRR, Precision@K

Feature Engineering:

·       Build features from user behavior, interaction signals, and context

·       Handle sparse and noisy data

Cold Start Handling:

·       Design strategies for new users and new offers

·       Implement hybrid and fallback approaches

Bias Handling:

·       Mitigate popularity and exposure bias

·       Implement diversity and re-ranking strategies

Model Evaluation:

·       Define and track CTR, conversion, NDCG, AUC

·       Continuously improve engagement metrics

Scalability & Deployment:

·       Build systems for millions of users

·       Enable real-time or near real-time inference

Required Skills

·       Strong experience in ML: classification, regression, propensity modeling

·       Experience with recommendation systems and ranking models

·       Hands-on with XGBoost, LightGBM

·       Python, SQL, feature engineering

·       Experience with MLOps and model deployment

Nice to Have

·       Experience with AWS SageMaker or Databricks

·       Experience with LLM-based recommendation approaches

·       A/B testing and experimentation knowledge

Role Summary

ML Engineer who can build a scalable offer recommendation system using propensity modeling and ranking techniques to improve user engagement in a dynamic, sparse-data environment.