--- title: Inside LinkedIn's data context layer: curating... description: LinkedIn describes a shared data context layer that helps both humans and AI agents discover trustworthy datasets and understand how to use them correctly. The... canonical: https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Inside LinkedIn's data context layer: curating trustworthy context for humans and agents | daily.dev og:description: LinkedIn describes a shared data context layer that helps both humans and AI agents discover trustworthy datasets and understand how to use them correctly. The... og:url: https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde og:image: https://api.daily.dev/og/posts/hL2pgQJdE.png og:image:alt: Inside LinkedIn's data context layer: curating trustworthy context for humans and agents og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Inside LinkedIn's data context layer: curating trustworthy context for humans and agents **[LinkedIn Engineering](https://daily.dev/sources/linkedin)** · 13 min read · 0 upvotes · 0 comments ## Summary LinkedIn describes a shared data context layer that helps both humans and AI agents discover trustworthy datasets and understand how to use them correctly. The system combines an AI enrichment pipeline that auto-generates dataset profiles from schemas, query logs, and source code with a human-curated layer where teams add business definitions, caveats, and known issues via Git pull requests. Context is organized into global, team, and analysis scopes, and agents navigate it through MCP tools using a hierarchical index system (progressive disclosure) to avoid loading everything into their context window. Internal results cite an acceptable-answer rate rising from 25.9% to 79.5% in one benchmark and median time-to-answer dropping from 10.7 to 4.7 minutes after integration. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does LinkedIn let AI agents retrieve relevant internal data context without overloading their context window? LinkedIn organizes context into a hierarchical tree of indexes, where each index holds a brief summary of its node plus a map of children to their summaries. Agents traverse this tree through a navigate team index MCP tool, reading summaries at each level to pick the right branch, then fetch the full document only once identified, typically in around four index reads. _Teams designing agent context retrieval can compare approaches like this one on daily.dev._ ### What results did LinkedIn see after integrating curated and generated dataset context into its analytics agent? In one business-area benchmark, the permission-adjusted acceptable-answer rate rose from 25.9% with the earlier analytics-agent experience to 79.5% after integrating the shared context layer. A separate comparison found the agent's median end-to-end time to answer a question fell from 10.7 to 4.7 minutes, suggesting gains in both answer quality and speed. _Anyone evaluating agent data-context strategies can track similar engineering case studies on daily.dev._ ### How does LinkedIn keep AI-generated dataset documentation from overwriting human-written corrections? Generated and curated context are stored separately but linked to the same asset identifier, so a single fetch returns both together. The enrichment pipeline only refreshes the generated layer, leaving human-curated corrections, exceptions, and usage guidance—authored through Git pull requests reviewed by asset owners—fully intact. _Engineers building data documentation pipelines can follow architecture patterns like this on daily.dev._ ## Similar posts on daily.dev - [Why AI agents need a Context Lake](https://daily.dev/posts/why-ai-agents-need-a-context-lake-bdkjprhe2) · The New Stack · 1 upvotes · 0 comments - [The Agent Context Layer for Trustworthy Data Agents](https://daily.dev/posts/the-agent-context-layer-for-trustworthy-data-agents-ksa3tmdsy) · Snowflake · 1 upvotes · 0 comments - [Data Context Is Not Enough: Why Trusted Data Context Wins](https://daily.dev/posts/data-context-is-not-enough-why-trusted-data-context-wins-sz2d9rimn) · Decube · 0 upvotes · 0 comments - [Why context is the biggest unlock for your AI strategy](https://daily.dev/posts/why-context-is-the-biggest-unlock-for-your-ai-strategy-vplu1mth2) · Atlassian · 1 upvotes · 0 comments - [We Kept Adding Context to Our AI Feature. It Kept Getting Worse](https://daily.dev/posts/we-kept-adding-context-to-our-ai-feature-it-kept-getting-worse-ahgwujixf) · Medium · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#linkedin](https://daily.dev/tags/linkedin) [View this post on daily.dev](https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Inside LinkedIn's data context layer: curating trustworthy context for humans and agents","url":"https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde"},"datePublished":"2026-10-09T07:52:07.947Z","dateModified":"2026-10-09T08:46:53.246Z","description":"LinkedIn describes a shared data context layer that helps both humans and AI agents discover trustworthy datasets and understand how to use them correctly. The...","image":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","thumbnailUrl":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","isAccessibleForFree":true,"articleSection":"LinkedIn Engineering","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"LinkedIn Engineering","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/le","url":"https://daily.dev/sources/linkedin"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,mcp,linkedin","timeRequired":"PT13M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"LinkedIn Engineering","item":"https://daily.dev/sources/linkedin"},{"@type":"ListItem","position":3,"name":"Inside LinkedIn's data context layer: curating trustworthy context for humans and agents"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/inside-linkedin-s-data-context-layer-curating-trustworthy-context-for-humans-and-agents-hl2pgqjde#faq","mainEntity":[{"@type":"Question","name":"How does LinkedIn let AI agents retrieve relevant internal data context without overloading their context window?","acceptedAnswer":{"@type":"Answer","text":"LinkedIn organizes context into a hierarchical tree of indexes, where each index holds a brief summary of its node plus a map of children to their summaries. Agents traverse this tree through a navigate team index MCP tool, reading summaries at each level to pick the right branch, then fetch the full document only once identified, typically in around four index reads. Teams designing agent context retrieval can compare approaches like this one on daily.dev."}},{"@type":"Question","name":"What results did LinkedIn see after integrating curated and generated dataset context into its analytics agent?","acceptedAnswer":{"@type":"Answer","text":"In one business-area benchmark, the permission-adjusted acceptable-answer rate rose from 25.9% with the earlier analytics-agent experience to 79.5% after integrating the shared context layer. A separate comparison found the agent's median end-to-end time to answer a question fell from 10.7 to 4.7 minutes, suggesting gains in both answer quality and speed. Anyone evaluating agent data-context strategies can track similar engineering case studies on daily.dev."}},{"@type":"Question","name":"How does LinkedIn keep AI-generated dataset documentation from overwriting human-written corrections?","acceptedAnswer":{"@type":"Answer","text":"Generated and curated context are stored separately but linked to the same asset identifier, so a single fetch returns both together. The enrichment pipeline only refreshes the generated layer, leaving human-curated corrections, exceptions, and usage guidance—authored through Git pull requests reviewed by asset owners—fully intact. Engineers building data documentation pipelines can follow architecture patterns like this on daily.dev."}}]} ```