Databricks Crowns AI Startup Innovators

Databricks announces VisionHeight as the grand prize winner of its 2026 Startup Challenge for its agentic threat intelligence platform.

4 min read
Databricks logo with trophy graphic and text '2026 Built-On Databricks Startup Challenge Winners'
Databricks celebrates the winners of its 2026 Built-On Databricks Startup Challenge.
Visual TL;DR
Databricks Startup ChallengeCore
annual competition for early-stage AI companies
From the article 3 mentionsDatabricks has announced the winners of its 2026 Built-On Databricks Startup Challenge, recognizing early-stage companies leveraging its Lakehouse platform for innovative AI solutions.
VisionHeight wins grand prizeOutcome
From the articleGrand prize winner VisionHeight impressed judges with its attack intelligence infrastructure, a novel approach to agentic threat intelligence.
Web Search AIEffect
From the article 2 mentionsSecond place went to Linkup for its production-grade Web Search API, designed to connect AI agents and chatbots with fresh, factual web data.
Retail AIEffect
other innovative AI solutions recognized
From the article 2 mentionsIntelo secured third place with its agentic workforce for retail merchandising and planning, unifying complex processes into a single intelligence layer.
Databricks LakehouseCore
platform used by winning startups
From the article 3 mentionsDatabricks has announced the winners of its 2026 Built-On Databricks Startup Challenge, recognizing early-stage companies leveraging its Lakehouse platform for innovative AI solutions.
Proactive CybersecurityContext
mapping adversary infrastructure before attacks launch
From the article 2 mentionsThe global competition saw startups across cybersecurity, retail, and web search present their core products.
AI Startup InnovatorsOutcome
Databricks crowns top AI startups
From the article 4 mentionsThe competition underscores the growing trend of startups building sophisticated AI applications on Databricks.
VisionHeight's platformCore
From the article 5 mentionsVisionHeight aims to close this gap by providing visibility into this staging process, allowing defenders to react before an attack materializes.

Databricks has announced the winners of its 2026 Built-On Databricks Startup Challenge, recognizing early-stage companies leveraging its Lakehouse platform for innovative AI solutions. The global competition saw startups across cybersecurity, retail, and web search present their core products.

Grand prize winner VisionHeight impressed judges with its attack intelligence infrastructure, a novel approach to agentic threat intelligence. The platform maps adversary infrastructure during its construction, offering a preemptive cybersecurity stance.

VisionHeight: Proactive Cybersecurity

Adversaries often build attack infrastructure weeks before launching campaigns. VisionHeight aims to close this gap by providing visibility into this staging process, allowing defenders to react before an attack materializes.

Web Search and Retail AI Shine

Second place went to Linkup for its production-grade Web Search API, designed to connect AI agents and chatbots with fresh, factual web data. Intelo secured third place with its agentic workforce for retail merchandising and planning, unifying complex processes into a single intelligence layer.

Other notable mentions included Clarecast for predictive intelligence, Gemini Sports for AI-powered squad planning in professional football, and LakeFusion for its AI-native data foundation platform.

Databricks as a Startup Launchpad

The competition underscores the growing trend of startups building sophisticated AI applications on Databricks. The company also announced an expanded startup program offering credits across Databricks and Neon to qualifying early-stage founders, aiming to accelerate their path from idea to market fit.

The winners were announced at the 2026 Data + AI Summit in San Francisco.

© 2026 StartupHub.ai. All rights reserved. Do not enter, scrape, copy, reproduce, or republish this article in whole or in part. Use as input to AI training, fine-tuning, retrieval-augmented generation, or any machine-learning system is prohibited without written license. Substantially-similar derivative works will be pursued to the fullest extent of applicable copyright, database, and computer-misuse laws. See our terms.