Machine Learning Engineer, Ranking & Retrieval

Role overview Build and operate machine learning systems that improve classification, retrieval, ranking, and personalization in AI-powered applications. You’ll take work from model development through production deployment, with an emphasis on measurable quality, responsiveness, cost, and reliability for real users.

Responsibilities - Develop and improve models for classification, ranking, retrieval, and personalization. - Improve search relevance and context retrieval using embeddings, hybrid search, and reranking approaches. - Create evaluation pipelines, labeled datasets, and metrics that track model quality and catch regressions. - Build and maintain reproducible training, inference, and deployment pipelines. - Monitor production performance, debug issues, and use real-world feedback to guide improvements. - Work with engineering and product colleagues to ship reliable ML-powered features while balancing accuracy, latency, and cost.

Requirements - Strong foundations in machine learning and modern neural architectures. - Hands-on experience deploying ML models in production. - Experience with one or more areas such as classification, ranking, retrieval, recommendation, or personalization. - Understanding of model evaluation, experimentation, and performance metrics. - Ability to write production-quality code and work with real-world datasets. - Sound judgment, ownership, and problem-solving skills, including balancing model quality, latency, cost, and reliability.

Nice to have - Experience with Python and PyTorch or JAX. - Familiarity with embeddings, vector search, ranking models, and production inference systems.

Benefits and work setup - Hands-on work in a small, high-talent-density team that makes decisions collaboratively and moves quickly. - A virtual and/or onsite interview process of three to four conversations; timing and work location are not specified.