--- title: Voice Agents are here. Context turns them into real... description: Databricks is pitching an architecture for enterprise voice agents that pairs a conversation layer with a governed 'brain' running on its platform. The brain... canonical: https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Voice Agents are here. Context turns them into real customer value. | daily.dev og:description: Databricks is pitching an architecture for enterprise voice agents that pairs a conversation layer with a governed 'brain' running on its platform. The brain... og:url: https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx og:image: https://api.daily.dev/og/posts/WjIQ3NAHx.png og:image:alt: Voice Agents are here. Context turns them into real customer value. 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. # Voice Agents are here. Context turns them into real customer value. **[databricks](https://daily.dev/sources/databricks)** · 10 min read · 1 upvotes · 0 comments ## Summary Databricks is pitching an architecture for enterprise voice agents that pairs a conversation layer with a governed 'brain' running on its platform. The brain looks up caller context, applies business rules, and picks the next step using both large language models for reasoning and fast 'System 1' decision models via Databricks' ai_decide feature, with Jev from TypeSafe AI cited as an example of such a model. A prototype uses LiveKit for call handling, Deepgram for speech-to-text, and ElevenLabs for voice. Databricks open-sourced a starting blueprint called DIVA on GitHub and recommends a 90-day pilot on one high-volume call type. Gartner is cited predicting agentic AI will resolve 80% of common customer service issues by 2029, cutting operational costs 30%. ## 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 System 1 and System 2 AI decision making in a voice agent? System 2 refers to large language models doing careful, slower reasoning, such as working out the right fix for a delayed order. System 1 refers to fast decision models, like Jev from TypeSafe AI, that handle quick calls such as deciding whether an upset caller should be routed to a human immediately. Databricks exposes System 1 decision making over governed data through a feature called ai_decide, which also returns a confidence score. _Teams weighing fast versus reasoning models for real-time agents can track these patterns on daily.dev._ ### What is the Databricks DIVA blueprint for building voice agents? DIVA, short for Databricks Intelligent Voice Agents, is an open source blueprint on GitHub meant as a working starting point for building a voice agent on Databricks. It includes ready-made instructions that coding agents can follow, and is intended to let a team pick one high-volume routine call, such as order status, and build toward production from there. _Developers evaluating open source starting points for voice agent projects can follow updates like this on daily.dev._ ### Which speech and voice technologies does Databricks use in its voice agent prototype? The prototype uses LiveKit to carry the call, Deepgram for speech-to-text and the default voice, and ElevenLabs for more expressive voices, with phone carriers like Twilio bringing in regular phone lines. Databricks itself provides the decision-making 'brain' that looks up context and chooses the agent's next step on each turn, separate from this conversation layer. _Anyone comparing speech-stack vendors for voice agent prototypes can follow these integrations on daily.dev._ ## Similar posts on daily.dev - [AI-Enabled Advisory Services for Higher Education](https://daily.dev/posts/ai-enabled-advisory-services-for-higher-education-wps7f1zel) · databricks · 0 upvotes · 0 comments - [From RAG to ontology: Databricks bets on context as the key to trusted AI agents](https://daily.dev/posts/from-rag-to-ontology-databricks-bets-on-context-as-the-key-to-trusted-ai-agents-ftjlhhcsd) · InfoWorld · 0 upvotes · 0 comments - [How AI Transformed Database Debugging at Databricks](https://daily.dev/posts/how-ai-transformed-database-debugging-at-databricks-xdhmgnzkw) · ByteByteGo · 8 upvotes · 0 comments - [I Built an AI Data Agent Which Can Query Data and Answer Business Questions. Here’s How.](https://daily.dev/posts/i-built-an-ai-data-agent-which-can-query-data-and-answer-business-questions-here-s-how--u3xbvpl4e) · Towards Data Science · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#llm](https://daily.dev/tags/llm), [#big-data](https://daily.dev/tags/big-data), [#databricks](https://daily.dev/tags/databricks), [#voice-ai](https://daily.dev/tags/voice-ai) [View this post on daily.dev](https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx) ```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":"Voice Agents are here. Context turns them into real customer value.","url":"https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx"},"datePublished":"2026-10-28T18:17:24.213Z","dateModified":"2026-10-28T18:25:18.165Z","description":"Databricks is pitching an architecture for enterprise voice agents that pairs a conversation layer with a governed 'brain' running on its platform. The brain...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4477197da2e244aa00b381e2a7a09c28?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4477197da2e244aa00b381e2a7a09c28?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"databricks","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":"databricks","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/fa7aa720f2db4d1eba826814730482c8","url":"https://daily.dev/sources/databricks"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,llm,big-data,databricks,voice-ai","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"databricks","item":"https://daily.dev/sources/databricks"},{"@type":"ListItem","position":3,"name":"Voice Agents are here. Context turns them into real customer value."}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/voice-agents-are-here-context-turns-them-into-real-customer-value--wjiq3nahx#faq","mainEntity":[{"@type":"Question","name":"What is the difference between System 1 and System 2 AI decision making in a voice agent?","acceptedAnswer":{"@type":"Answer","text":"System 2 refers to large language models doing careful, slower reasoning, such as working out the right fix for a delayed order. System 1 refers to fast decision models, like Jev from TypeSafe AI, that handle quick calls such as deciding whether an upset caller should be routed to a human immediately. Databricks exposes System 1 decision making over governed data through a feature called ai_decide, which also returns a confidence score. Teams weighing fast versus reasoning models for real-time agents can track these patterns on daily.dev."}},{"@type":"Question","name":"What is the Databricks DIVA blueprint for building voice agents?","acceptedAnswer":{"@type":"Answer","text":"DIVA, short for Databricks Intelligent Voice Agents, is an open source blueprint on GitHub meant as a working starting point for building a voice agent on Databricks. It includes ready-made instructions that coding agents can follow, and is intended to let a team pick one high-volume routine call, such as order status, and build toward production from there. Developers evaluating open source starting points for voice agent projects can follow updates like this on daily.dev."}},{"@type":"Question","name":"Which speech and voice technologies does Databricks use in its voice agent prototype?","acceptedAnswer":{"@type":"Answer","text":"The prototype uses LiveKit to carry the call, Deepgram for speech-to-text and the default voice, and ElevenLabs for more expressive voices, with phone carriers like Twilio bringing in regular phone lines. Databricks itself provides the decision-making 'brain' that looks up context and chooses the agent's next step on each turn, separate from this conversation layer. Anyone comparing speech-stack vendors for voice agent prototypes can follow these integrations on daily.dev."}}]} ```