Lead Applied Scientist, Search & Information Retrieval
Core
Design, develop, and production deploy large-scale search and information retrieval systems for legal, tax, and accounting content.
Role type
Lead Applied Scientist, Search & Information Retrieval
Builds
Search architectures, indexing pipelines, ranking/re-ranking systems, and self-service search platforms for Thomson Reuters products.
Domain
Legal, tax, and accounting technology; Information Retrieval and Machine Learning.
Deliverable
production ML models
Required skills
Search engine architecture, indexing systems, ingestion pipelines, ranking and re-ranking systems, information retrieval, semantic retrieval, hybrid retrieval, vector search, query understanding, relevance optimization, LLM-enhanced retrieval, RAG architectures, end-to-end measurement and evaluation of search quality.
Preferred skills
Legal/regulatory/tax/scientific domain experience, enterprise knowledge repositories, Elasticsearch/OpenSearch/Solr/Vespa, API platform development, Agentic AI systems, AzureML/AWS SageMaker, document understanding.
Technologies
Python, PyTorch, Hugging Face Transformers, Elasticsearch, OpenSearch, Solr, Vespa, AzureML, AWS SageMaker.
Responsibilities
Design and deploy search architectures for large-scale content collections; Build and optimize ingestion pipelines; Develop ranking and re-ranking systems using traditional IR and LLM-based approaches; Improve retrieval quality through semantic and hybrid retrieval; Design evaluation frameworks for retrieval performance; Lead technical decisions on indexing and search infrastructure; Partner with engineering teams to deliver scalable services; Mentor applied scientists and ML practitioners.
Seniority
Senior, hands-on IC with leadership responsibilities.