Principal Applied Scientist
Core
Advance retrieval quality science through evaluation frameworks and metrics for multi-hop retrieval, cross-source reasoning, semantic enrichment, knowledge graphs, and agent memory.
Role type
Principal Applied Scientist (Enterprise AI & Retrieval)
Builds
Trust-aware AI systems, autonomous knowledge acquisition pipelines, and production-grade retrieval agents at hyperscale.
Domain
Enterprise AI, Knowledge Graphs, Retrieval Systems, Agentic Systems
Deliverable
production ML models
Required skills
Multi-hop retrieval, cross-source reasoning, semantic enrichment, knowledge graph generation, agent memory, evaluation framework design, quality metrics definition, enterprise AI architecture, platform architecture influence, cross-team roadmap alignment
Preferred skills
None explicitly stated
Technologies
SEVALs, Recall@K, grounding quality, citation correctness, freshness, coverage, task success
Responsibilities
Develop techniques for multi-hop retrieval and cross-source reasoning; Build trust-aware AI systems respecting enterprise permissions; Pioneer autonomous knowledge acquisition including relationship discovery and summarization; Partner with engineering and product teams to bring scientific innovations into production; Define evaluation frameworks and quality metrics for Agentic systems
Seniority
Principal, hands-on IC with strategy & mentorship
