//covers data governance, information architecture, enterprise data management
New York, USAAtlan editorial contributorsince Jan 2026
Emily Winks is a data governance expert who came to data through library and information science. Over 18+ years she has moved from children's librarian to information architect at WeWork to Atlan's Founder's Office, bringing a cataloging discipline to how enterprises classify, govern, and trust their data. She writes on data governance, enterprise data management, and the context layer.
/ Profile
About Emily.
After graduating with a Master of Library and Information Science and a certificate in Records Management and Archives, Emily began her career as a children's librarian. After seven years in libraries she left to join the corporate workforce — she doesn't sing The Hokey Pokey (as often) anymore, but still puts her whole self in when it comes to navigating the ambiguities of data, making digital information clear, and empowering others to do the same.
She spent her library years at the Smithtown Special Library District and as a Senior Librarian at the New York Public Library before moving into information architecture. At WeWork she progressed from Information Architect to Senior Information Architect to Data Product Architect, building the classification and data structures behind a fast-scaling business.
Emily joined Atlan's Founder's Office in 2023, where she works at the intersection of data governance, enterprise data management, and the context layer — writing on how organizations classify, govern, and make their data trustworthy for both teams and AI agents.
Prompt engineering, context engineering, harness engineering: three disciplines, three failure modes. Learn where each layer breaks down and how Atlan's Context Layer ships with Context…
Updated Jul 21, 2026·Published Apr 13, 2026·20 min read
A data lakehouse for AI combines reliable storage with governed metadata. Atlan adds certified definitions, lineage, and policy so AI agents trust the data.
Building a knowledge base for AI agents takes five layers: ingestion, hybrid retrieval, reranking, evaluation, and a governed semantic layer. Here's how.
How to connect enterprise data sources to LLMs securely: enforce access control, mask PII, manage secrets, and meet residency rules at every connection.
Learn how to handle PII in AI pipelines: discovery, dynamic masking, lineage, and MCP-gated access that keep sensitive data safe before it reaches a model.
Real-time data for AI agents needs more than fast streaming; agents need context freshness, lineage, and policy state before they can trust an event and…
Structured vs. unstructured data for AI creates a context gap: agents need both, unified by shared meaning. See how a governed context layer closes it.
Emily is a member of the Atlan team. Articles published under her byline reflect her own analysis and expertise; Atlan's editorial team works with her on structure and clarity but does not author content attributed to her. She reviews and approves every piece published under her byline before publication.
