--- title: Context, Semantics, and Ontology: A Primer for the... description: A long-form primer distinguishing three overlapping concepts in modern data architecture: the semantic layer (deterministic, SQL-compiled metric definitions... canonical: https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Context, Semantics, and Ontology: A Primer for the Agentic Era | daily.dev og:description: A long-form primer distinguishing three overlapping concepts in modern data architecture: the semantic layer (deterministic, SQL-compiled metric definitions... og:url: https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte og:image: https://api.daily.dev/og/posts/kWvuTpNtE.png og:image:alt: Context, Semantics, and Ontology: A Primer for the Agentic Era 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. # Context, Semantics, and Ontology: A Primer for the Agentic Era **[MotherDuck](https://daily.dev/sources/motherduck)** · 19 min read · 6 upvotes · 0 comments ## Summary A long-form primer distinguishing three overlapping concepts in modern data architecture: the semantic layer (deterministic, SQL-compiled metric definitions like Cube, Malloy, dbt Semantic Layer), the context layer (curated unstructured business knowledge such as Notion docs and Jira tickets that agents ingest), and ontologies (schema-level world models like Palantir Foundry's object/link/property types). The piece argues these emerged because AI agents can now digest unstructured context that was previously unmanageable, creating a paradox where agents both necessitate and enable richer context. It discusses the new Apache Ossie (formerly Open Semantic Interchange) standard for vendor-agnostic semantic model definitions, the lack of a data-quality scoring standard for context, and contrasts graph-based retrieval versus flat-index approaches (citing MotherDuck's Guides feature and Anthropic's semantic-layer-first Claude skill). The conclusion: tools aren't competing, they're complementary, and human judgment remains essential for deciding what data actually means and keeping context correct over time. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### what is the difference between a semantic layer and a context layer in data engineering A semantic layer defines logic that compiles deterministically to SQL, such as metrics and dimensions, and sits between business users and a database so queries use familiar business concepts. A context layer instead holds unstructured, free-flowing information like Notion docs, Confluence pages, and past query history that agents use for reasoning, without the same deterministic SQL guarantee. _daily.dev surfaces practical breakdowns like this for teams weighing semantic versus context layer investments._ ### what is the Open Semantic Interchange OSI standard for semantic layers Open Semantic Interchange, recently renamed Apache Ossie, is an open-source, vendor-agnostic specification that standardizes the syntax for semantic layer declarative configuration. It aims to avoid the proliferation of proprietary YAML formats used by tools like LookML, Cube, dbt Semantic Layer, and Snowflake's internal semantic layer, letting a single definition map to any semantic layer implementation. _following standards like Apache Ossie on daily.dev helps teams avoid lock-in when building semantic layers._ ### why does Anthropic's Claude skill for data analytics use the semantic layer as the mandatory default over raw SQL Anthropic's skill instructions require agents to try the semantic layer first for every data question, falling back to raw SQL only once the semantic-layer path is shown not to cover the request. This ordering exploits the semantic layer's deterministic, governed definitions of metrics and joins before letting the agent write ad hoc SQL that risks misinterpreting business logic like revenue definitions. _daily.dev keeps builders informed on how teams like Anthropic sequence semantic layers and SQL for agent workflows._ ## Similar posts on daily.dev - [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 --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#data-engineering](https://daily.dev/tags/data-engineering) [View this post on daily.dev](https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte) ```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":"Context, Semantics, and Ontology: A Primer for the Agentic Era","url":"https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte"},"datePublished":"2026-09-02T14:37:41.332Z","dateModified":"2026-09-14T09:11:46.351Z","description":"A long-form primer distinguishing three overlapping concepts in modern data architecture: the semantic layer (deterministic, SQL-compiled metric definitions...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f07f1c8ff829f719cc17100933670dbd?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f07f1c8ff829f719cc17100933670dbd?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"MotherDuck","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":"MotherDuck","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/f084d6835eef4c2b83f889f9a4e85ed6","url":"https://daily.dev/sources/motherduck"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":6},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,data-engineering","timeRequired":"PT19M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"MotherDuck","item":"https://daily.dev/sources/motherduck"},{"@type":"ListItem","position":3,"name":"Context, Semantics, and Ontology: A Primer for the Agentic Era"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/context-semantics-and-ontology-a-primer-for-the-agentic-era-kwvutpnte#faq","mainEntity":[{"@type":"Question","name":"what is the difference between a semantic layer and a context layer in data engineering","acceptedAnswer":{"@type":"Answer","text":"A semantic layer defines logic that compiles deterministically to SQL, such as metrics and dimensions, and sits between business users and a database so queries use familiar business concepts. A context layer instead holds unstructured, free-flowing information like Notion docs, Confluence pages, and past query history that agents use for reasoning, without the same deterministic SQL guarantee. daily.dev surfaces practical breakdowns like this for teams weighing semantic versus context layer investments."}},{"@type":"Question","name":"what is the Open Semantic Interchange OSI standard for semantic layers","acceptedAnswer":{"@type":"Answer","text":"Open Semantic Interchange, recently renamed Apache Ossie, is an open-source, vendor-agnostic specification that standardizes the syntax for semantic layer declarative configuration. It aims to avoid the proliferation of proprietary YAML formats used by tools like LookML, Cube, dbt Semantic Layer, and Snowflake's internal semantic layer, letting a single definition map to any semantic layer implementation. following standards like Apache Ossie on daily.dev helps teams avoid lock-in when building semantic layers."}},{"@type":"Question","name":"why does Anthropic's Claude skill for data analytics use the semantic layer as the mandatory default over raw SQL","acceptedAnswer":{"@type":"Answer","text":"Anthropic's skill instructions require agents to try the semantic layer first for every data question, falling back to raw SQL only once the semantic-layer path is shown not to cover the request. This ordering exploits the semantic layer's deterministic, governed definitions of metrics and joins before letting the agent write ad hoc SQL that risks misinterpreting business logic like revenue definitions. daily.dev keeps builders informed on how teams like Anthropic sequence semantic layers and SQL for agent workflows."}}]} ```