Sr. Associate Knowledge Manager - 11609

What You'll Do:

• Prepare existing knowledge systems (e.g., Confluence, Highspot) for Retrieval-Augmented Generation (RAG). Clean, deduplicate, and format unstructured content to ensure AI agents ingest only pristine, non-contradictory data. 
• Certify data sources prior to IT development by enforcing strict metadata labeling, aggressively archiving outdated content, and establishing bounded contexts so LLMs do not hallucinate on legacy garbage. 
• Convert complex human-readable formats (nested macros, embedded PDFs, poorly structured tables) into clean, standard formats (like Markdown) that can be easily parsed by AI models. 
• Enforce Golden Data standards. Act as the ruthless gatekeeper for knowledge repositories used by Value Services. Partner with stakeholders to ensure content strictly adheres to "Single Source of Truth" principles before it is approved for agent ingestion. 
• Establish reporting to identify "data rot" (stale pages, broken links, contradictory policies) and actively remediate it before it impacts AI agent performance.
• Act as a strategic liaison to the enterprise Data, AI, & Analytics organization to align local content standards with corporate policies.

What You Will Bring to Coupa:

• 3-5 years having led large-scale content auditing, deduplication, and migration initiatives, transforming high volumes of unstructured documentation into clean, governed, and AI-ready knowledge repositories.
• Designed and implemented metadata frameworks, taxonomy structures, and relational data models to convert unstructured content into structured, searchable, and scalable information architectures.
• Applied deep understanding of large language models (LLMs) to optimize content formatting, labeling, and governance practices, improving AI accuracy and reducing hallucination risks.
• Administered and optimized enterprise knowledge management platforms including Confluence and Highspot, leveraging Markdown and HTML to standardize content organization and accessibility.
• Established and enforced data governance standards, taxonomy conventions, and content quality controls across cross-functional teams without reliance on enterprise taxonomy management tools.
• Utilized a critical "Red Team" approach to evaluate data quality, identify gaps, bias, and inaccuracies, isolating high-value content and ensuring reliable AI reasoning, knowledge retrieval, and decision support outcomes.