About this role
Own applied ML work end-to-end for recommendation system workloads, adapting Liquid Foundation Models for personalization and ranking while building reusable applied tooling. Design and execute large-scale data pipelines, fine-tune sequential recommendation models (e.g., HSTU-style), and design task-specific evaluations for ranking quality, latency, and throughput for enterprise customers.
Skills for this role
Recommendation SystemsSequential RecommendationRanking SystemsUser Behavior ModelingFeature EngineeringData Pipeline DesignTraining Data CurationModel Fine-tuningHSTUSASRecBERT4RecPythonPyTorchEvaluation DesignA/B TestingLatency OptimizationThroughput OptimizationModel ServingData QualityTooling and Workflow DevelopmentCustomer-facing Delivery