Principal Applied Scientist
Core
Design and run experiments to develop scalable retrieval pipelines and models for enterprise-scale search systems, including semantic, lexical, and hybrid retrieval, RAG, and agentic AI architectures.
Role type
Principal Applied Scientist (Retrieval Science)
Builds
Enterprise-scale search systems, retrieval pipelines, and LLM-integrated retrieval architectures
Domain
Search, Information Retrieval, Large Language Models (LLMs), Agentic AI
Deliverable
production ML models
Required skills
Semantic retrieval (late-interaction architectures), Dense retrieval, Embedding model fine-tuning, SPLADE/sparse retrieval, Hybrid retrieval systems, Query understanding, Multi-stage ranking, RAG, Reinforcement learning, LLM-integrated architectures, Python, PyTorch, Offline/online evaluation metrics, Scalability optimization
Preferred skills
Publications in top-tier conferences (SIGIR, RecSys, KDD, WWW, WSDM, ACL, EMNLP), Experience with agentic AI systems, Personalization and contextual ranking
Technologies
ColBERT, SPLADE, PyTorch, Python, LLMs
Responsibilities
Develop scalable retrieval pipelines and models, Define offline and online evaluation metrics, Provide technical leadership and mentorship, Establish and evolve evaluation frameworks for retrieval quality and Copilot metrics
Seniority
Principal, hands-on IC with mentorship
