As a Machine Learning Engineer specializing in Recommender Systems, you will design, build, and optimize large-scale recommendation architectures that power personalized user experiences for enterprise environments. You’ll work across candidate generation, deep learning ranking models, and high-throughput real-time systems, owning recommendation pipelines end-to-end from experimentation to production deployment.

Functional Responsibilities:

• Design and implement multi-stage recommendation pipelines, including Candidate Generation (Retrieval), Ranking, and Re-ranking/Filtering stages.
• Develop and fine-tune machine learning and deep learning models for personalized recommendations using techniques such as Collaborative Filtering, Matrix Factorization, Two-Tower Networks, and Deep Learning (e.g., Deep & Cross Networks).
• Build and optimize two-stage retrieval architectures using Approximate Nearest Neighbors (ANN) vector search engines (e.g., Faiss, Pinecone, Milvus).
• Implement real-time scoring and inference pipelines using feature stores (e.g., Feast, Hopsworks) and scalable serving frameworks (e.g., Triton, TorchServe, Ray Serve).
• Establish A/B testing frameworks, offline evaluation metrics (NDCG, MAP, Recall@K), and real-time monitoring for model performance and business metrics.
• Optimize recommendation system latency and throughput using quantization, caching, and hardware acceleration.

Qualifications:

• 5+ years of hands-on experience in machine learning and software engineering, with proven experience building and deploying large-scale recommender systems in production.
• Advanced English proficiency (written and spoken) with strong communication skills to articulate technical recommendations to cross-functional stakeholders.
• Strong Python programming skills with expertise in machine learning frameworks (PyTorch, TensorFlow) and recommendation libraries (e.g., Surprise, LightFM, Implicit, RecBole, NVIDIA Merlin).
• Demonstrated experience with multi-stage recommendation techniques, vector databases/ANN search (Faiss, Milvus, Pinecone), feature stores, and modern serving frameworks.
• Solid background in cloud platforms (AWS, GCP, Azure), MLOps pipelines, SQL, and big data processing tools (Apache Spark) for feature engineering at scale.