Kana emerges from stealth with $15M to build flexible AI agents for marketers

Kana raises 15 million dollars

San Francisco-based startup Kana launched out of stealth today with $15 million in seed funding to automate marketing operations through autonomous AI agents. Mayfield led the investment round, which marks the fourth venture for co-founders Tom Chavez and Vivek Vaidya. The company spent nine months in incubation at super{set}, a startup studio also founded by Chavez and Vaidya. The platform enters a crowded market where incumbents like Google, Meta, and Microsoft already offer automated tools for advertisers. Kana attempts to differentiate itself by deploying specialized AI agents that handle specific tasks across the marketing lifecycle. These agents manage data analysis, audience targeting, and media planning while integrating directly with existing legacy software. Navin Chaddha, managing partner at Mayfield, will join Kana’s board of directors as part of the deal. The company plans to use the new capital to expand its engineering and product teams. It will also scale its go-to-market strategy to reach enterprise clients currently struggling with fragmented marketing stacks.

Experienced founders tackle marketing debt

CEO Tom Chavez and CTO Vivek Vaidya bring 25 years of experience in marketing technology to this new venture. Their track record includes several high-profile exits to major tech players over the last two decades. This history gives the startup immediate credibility in a sector often wary of unproven AI tools. Their previous companies focused on data management and optimization long before the current generative AI surge. Rapt, their first venture, focused on media monetization and was acquired by Microsoft in 2008. Their second major success, Krux, specialized in data management and sold to Salesforce in 2016 for approximately $700 million. The founders spent nearly a year developing Kana within their own startup studio, super{set}. They claim this period allowed them to test the agentic architecture against real-world marketing pain points. Chavez argues that large companies have "wallowed" in inefficient systems for too long, creating an opening for a more flexible, AI-native approach.

Autonomous agents manage complex workflows

Kana’s platform relies on a series of "loosely coupled" AI agents designed to perform distinct roles within a marketing department. Unlike monolithic software suites, these agents can be reconfigured or swapped out depending on the specific needs of a campaign. They operate simultaneously to streamline tasks that usually require multiple human teams or disconnected software tools. A typical workflow on the platform begins with a media brief. The AI agents analyze the document to extract goals, identify target demographics, and pull market research data. This process replaces manual data entry and cross-referencing between different research platforms. The system includes several core capabilities for enterprise marketers:

• Data Analysis: Agents scan internal and external datasets to find patterns in consumer behavior.
• Audience Targeting: The platform identifies specific segments most likely to convert based on historical campaign performance.
• Campaign Management: Autonomous agents monitor active ads and adjust spending or creative assets in real-time.
• Media Planning: The software suggests budget allocations across different platforms to maximize return on investment.
• Chatbot Optimization: Agents tune customer-facing AI to ensure messaging aligns with broader marketing goals.

Synthetic data reduces reliance on third parties

One of Kana’s most distinct features is its ability to generate synthetic data for market research. This technology creates artificial datasets that mirror the statistical properties of real consumer groups without compromising individual privacy. Marketers use this to fill gaps in their own data or to augment expensive third-party sources. Chavez notes that relying on external data providers has become increasingly expensive and legally complex due to privacy regulations. Synthetic data allows companies to run simulations and test marketing strategies before spending money on actual ad placements. This "dry run" capability helps narrow down strategies and reduces the overall cost of experimentation. The platform also uses this data to improve audience targeting. By creating synthetic profiles that represent ideal customers, the AI can better predict how real users will respond to specific messaging. This approach aims to solve the "cold start" problem where new brands lack enough historical data to train effective AI models.

Human oversight remains a core requirement

Despite the focus on autonomy, Kana keeps humans in the loop to approve and refine agent actions. Marketers can review the decisions made by the AI, provide feedback, and customize the parameters of each task. This hybrid approach prevents the AI from making costly errors without human supervision. The founders emphasize that flexibility is their primary advantage over established competitors. While legacy systems often require months of integration and training, Kana’s agents can be deployed and tailored in real-time. This allows marketing teams to see results on their campaigns faster than they would with traditional enterprise software. Vaidya describes this philosophy as "build with" rather than simply "buy." Instead of selling a rigid product, Kana provides a framework that companies can configure to meet their specific operational needs. This customization serves as a moat against larger tech firms that typically offer one-size-fits-all solutions.

The competitive landscape for AI marketing

The launch comes at a time when the marketing industry is saturated with AI startups like Jasper and Copy.ai. Most of these early players focused on content generation, such as writing blog posts or social media captions. Kana is moving further "upstream" by focusing on the underlying data and strategy that dictates what content gets made in the first place. Incumbents are also moving fast to protect their territory. Salesforce recently integrated its Einstein GPT across its marketing cloud, and Google continues to bake AI directly into its Ads platform. Kana’s leadership believes these big-tech solutions are too rigid for the modern marketer who needs to move across different platforms and ecosystems. The $15 million in seed funding will support a hiring push across three main departments. Kana is currently looking for specialists in engineering, product design, and go-to-market execution. By scaling its team, the startup hopes to prove that its agentic model can outperform the "black box" automation currently offered by the world's largest advertising platforms.