--- title: Enterprise Agents Have a Structure Problem - Ishita... description: A Tesla ML engineer presents a framework for diagnosing why enterprise data agents fail, identifying three core structural problems: ambiguity (agents lack a... canonical: https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla | daily.dev og:description: A Tesla ML engineer presents a framework for diagnosing why enterprise data agents fail, identifying three core structural problems: ambiguity (agents lack a... og:url: https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm og:image: https://api.daily.dev/og/posts/b6nbbV8LM.png og:image:alt: Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla 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. # Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 12 min read · 0 upvotes · 0 comments ## Summary A Tesla ML engineer presents a framework for diagnosing why enterprise data agents fail, identifying three core structural problems: ambiguity (agents lack a clear hierarchy for choosing knowledge sources), staleness (context becomes outdated without a feedback loop), and preference (individual/team-level metric preferences are hard to capture). The proposed solutions include a three-tier source-of-truth hierarchy (semantic layer, canonical tables, database graph), a context lifecycle with live data sources and evaluation feedback loops, and partial approaches to preference capture via semantic layers and agent memory — though preference remains an open research problem. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why do enterprise AI data agents give wrong answers even with a bigger or newer model? Bigger models and more context do not fix the underlying structural issues. Enterprise agents commonly fail due to three problems: ambiguity over which data source or knowledge base holds the truth, staleness from outdated context like deprecated metric definitions, and unresolved team or individual preferences for how a metric should be calculated. _Teams debugging agent reliability issues can track structural fixes like this one on daily.dev._ ### What is the recommended hierarchy of sources of truth for a data agent to reduce ambiguity? A three-tier hierarchy works best: a semantic layer of curated KPI and metric definitions as the cleanest source, canonical parametric tables as a more flexible middle layer, and a full database graph connecting tables, columns, and metrics as the most flexible but hardest to maintain option. Starting with the first two layers reportedly solves about 80% of problems. _Engineers designing agent data architecture can follow discussions like this on daily.dev._ ### How can a team keep an enterprise AI agent's context from going stale? Build a context lifecycle with two components: embedding live data sources that are continuously updated and reviewed (like GitHub, CRM tools, Tableau, or DBT semantic layers), and a feedback loop that logs every correction event, such as an incorrect definition or a new filter, then uses human or automated evaluation to update the agent's context regularly. _Practitioners maintaining agent context pipelines can follow this kind of guidance on daily.dev._ ## 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 - [The Intelligible World of Agents](https://daily.dev/posts/the-intelligible-world-of-agents-bazldoigc) · Recorded Future Blog · 0 upvotes · 0 comments - [5 data foundation and technology stack gaps stalling your AI agents](https://daily.dev/posts/5-data-foundation-and-technology-stack-gaps-stalling-your-ai-agents-l9rkmz54j) · elastic · 1 upvotes · 0 comments - [Connecting AI agents to enterprise knowledge](https://daily.dev/posts/connecting-ai-agents-to-enterprise-knowledge-wjpvnq4xg) · MIT Technology Review · 0 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#ai-agents](https://daily.dev/tags/ai-agents), [#data-engineering](https://daily.dev/tags/data-engineering), [#context-engineering](https://daily.dev/tags/context-engineering) [View this post on daily.dev](https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm) ```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":"Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla","url":"https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm"},"datePublished":"2026-07-20T06:39:36.789Z","dateModified":"2026-09-14T08:32:48.137Z","description":"A Tesla ML engineer presents a framework for diagnosing why enterprise data agents fail, identifying three core structural problems: ambiguity (agents lack a...","image":"https://i.ytimg.com/vi/B8l81jhvHbI/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/B8l81jhvHbI/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","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":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,ai-agents,data-engineering,context-engineering","timeRequired":"PT12M","video":{"@type":"VideoObject","name":"Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla","description":"A Tesla ML engineer presents a framework for diagnosing why enterprise data agents fail, identifying three core structural problems: ambiguity (agents lack a...","thumbnailUrl":"https://i.ytimg.com/vi/B8l81jhvHbI/sddefault.jpg","uploadDate":"2026-07-20T06:39:36.789Z","duration":"PT12M","url":"https://api.daily.dev/r/b6nbbV8LM","embedUrl":"https://www.youtube.com/embed/B8l81jhvHbI"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"Enterprise Agents Have a Structure Problem - Ishita Daga, Tesla"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/enterprise-agents-have-a-structure-problem---ishita-daga-tesla-b6nbbv8lm#faq","mainEntity":[{"@type":"Question","name":"Why do enterprise AI data agents give wrong answers even with a bigger or newer model?","acceptedAnswer":{"@type":"Answer","text":"Bigger models and more context do not fix the underlying structural issues. Enterprise agents commonly fail due to three problems: ambiguity over which data source or knowledge base holds the truth, staleness from outdated context like deprecated metric definitions, and unresolved team or individual preferences for how a metric should be calculated. Teams debugging agent reliability issues can track structural fixes like this one on daily.dev."}},{"@type":"Question","name":"What is the recommended hierarchy of sources of truth for a data agent to reduce ambiguity?","acceptedAnswer":{"@type":"Answer","text":"A three-tier hierarchy works best: a semantic layer of curated KPI and metric definitions as the cleanest source, canonical parametric tables as a more flexible middle layer, and a full database graph connecting tables, columns, and metrics as the most flexible but hardest to maintain option. Starting with the first two layers reportedly solves about 80% of problems. Engineers designing agent data architecture can follow discussions like this on daily.dev."}},{"@type":"Question","name":"How can a team keep an enterprise AI agent's context from going stale?","acceptedAnswer":{"@type":"Answer","text":"Build a context lifecycle with two components: embedding live data sources that are continuously updated and reviewed (like GitHub, CRM tools, Tableau, or DBT semantic layers), and a feedback loop that logs every correction event, such as an incorrect definition or a new filter, then uses human or automated evaluation to update the agent's context regularly. Practitioners maintaining agent context pipelines can follow this kind of guidance on daily.dev."}}]} ```