Senior Machine Learning Engineer

About the role

A Senior Machine Learning Engineer designs, builds and scales production-grade algorithmic solutions for a digital commerce platform, working at the intersection of deep learning, reinforcement learning and distributed computing. The role covers real-time dynamic pricing and personalization, generative AI travel assistants and demand forecasting, and involves close collaboration with product and engineering teams to deploy low-latency, high-throughput models serving millions of users.

Key responsibilities

• Design and implement reinforcement learning and contextual bandit algorithms for real-time pricing, personalized discounts and multi-product bundling.
• Develop deep learning architectures such as Two-Tower models, Graph Neural Networks and Transformers to rank offerings, inventory and experiences.
• Build and fine-tune Large Language Models to power AI assistants for itinerary planning and customer support.
• Construct spatio-temporal forecasting algorithms for demand surges, cancellation probabilities and market competitiveness.
• Optimize models for millisecond-level inference during peak traffic.
• Architect and scale distributed training pipelines with PyTorch, Ray or DeepSpeed over billions of clickstream events and transactions.
• Implement evaluation frameworks, multi-variant A/B testing and counterfactual reasoning to measure causal impact on conversion rates and Gross Bookings Value.

Required profile

• 8+ years of experience building, deploying and scaling advanced machine learning algorithms in a high-throughput, customer-facing digital economy.
• Master's or Ph.D. in Computer Science, Operations Research, Applied Mathematics, Data Science or a highly quantitative field.
• Proven track record in at least two of: learning-to-rank and collaborative filtering; multi-armed bandits and reinforcement learning; large-scale time-series forecasting; big data and cloud infrastructure.

Required skills

• Expert Python or SQL, with production-grade, optimized code.
• PyTorch, TensorFlow/JAX, Scikit-Learn, XGBoost, LightGBM.
• Spark, Kafka/Flink, Airflow and cloud platforms (AWS, GCP or Azure).