AI vendors promote their enterprise products as if they were turnkey solutions, but the chances of AI agents going live right away are low. Unless you put in the effort to train a model on the specifics of your business, you’re unlikely to understand how your company, for example, defines revenue or know who has permission to see which file. That’s part of the reason we see AI companies deploying engineers to help integrate their AI products into customers’ systems.
New York-based startup jedify is attacking this same gap. The company says its platform connects to companies’ knowledge sources via APIs to create a “context graph” about their business that AI agents can use to work better. These sources can be databases, data lakes and warehouses, SaaS applications or BI tools, as well as unstructured sources such as reports, documentation, code bases, and even Slack channels and meeting recordings.
To achieve this, Khedify has raised $24 million in a Series A funding round led by Norwest, TechCrunch has learned exclusively. The round featured participation from returning backers S Capital VC and Cerca Partners, as well as new investor Oceans Ventures. Data giant Snowflake also participated as a strategic investor and is integrating the startup’s technology with its artificial intelligence products, such as its Cortex AI service, Semantic Views and CoWork.
Khedify’s argument is that to be useful within enterprises, AI agents need access to entity relationships, data, permissions, domain knowledge, workflows, operational assumptions, and enterprise-specific terminology. This context, the company says, allows an AI agent to limit its attention to information that is relevant to a particular task rather than searching through everything a company has.
Co-founder and CEO Assaf Henkin (pictured above, far right) pointed to Kiteworks, a fulfillment company, as an example of how customers use Khedify. Kiteworks connected Snowflake, Tableau, Notion, and internal guides, including documents and screenshots, to Jedify, and then created agency tools for different client workflows.
“They wanted to equip their salespeople and account teams with a sophisticated app; you can think of it as a dashboard app and a real-time conversation app. When they get into a conversation with a customer, Khedify builds for them, on the fly, everything they need to know. And during the conversation, they can, in real time, get very specific details that come up proactively,” Henkin said.
Henkin argues that Jedify’s context graph is different from the semantic layers, metadata catalogs, and knowledge graphs that companies already use because it is multidimensional and captures relationships between entities, data, people, permissions, and customers. It is also model agnostic and updates in real time as information enters and leaves the systems to which it is connected.
“When you want to enable an agent solution to be truly autonomous, to drive decisions through CRM data, Zendesk tickets, maybe telemetry data coming in in real time, that’s when a context graph is much better in terms of capabilities than a semantic layer,” he said.
Permits are an obvious hurdle here. For example, it wouldn’t be wise for an agent to give an intern access to the CFO’s revenue projections. Henkin said its platform works to address this by inheriting permissions from identity systems, file systems, SaaS tools and databases, including access rules at the row, column and table level, and then allowing its customers to create additional groups that define what and who agents or workflows are allowed to reach. It also offers observability and governance tools to help customers ensure their AI agents behave as intended.
Currently, Khedify targets medium to large enterprise customers who have mature data stacks and multiple databases or data warehouses. Henkin said the company has between 10 and 20 initial customers, one of which is The Weather Company, and is seeing interest from data-intensive sectors such as gaming, industrial and consumer packaged goods.
Snowflake’s investment and partnership is notable because big data platforms are also trying to build similar capabilities. But Henkin maintains that Khedify is complementary to such efforts because much of a company’s data, and most of its institutional knowledge, is typically not stored in a single cloud provider.
“(Big data companies) will tell you, ‘Oh yeah, bring it all in.’ But in reality, companies have multiple databases, warehouses and data solutions (…) The important thing is that not all of their data is in those environments, and most of their knowledge is not there, so it’s a bit of a disadvantage that they actually have,” he said.
Henkin also noted that for companies trying to do this on their own, training an AI model to build a comparable context layer can be cost-prohibitive, especially because Companies are examining and clamping down on the use of AI tokens.
And rapid advances in the development of AI models contribute to the company’s broader bet: As models become more capable and more interchangeable, the proprietary context that helps those models work better within companies could prove a valuable and lasting moat.
The startup will use the fresh money for product development, hiring and marketing. This brings the company’s total funding to about $33 million.
When you purchase through links in our articles, we may earn a small commission. This does not affect our editorial independence.
