Jedify raises $24M to provide enterprise AI agents with missing business context

Jedify Inc., an enterprise artificial intelligence context startup, today announced that it has raised $24 million in new funding to build what it calls a context graph, which provides the business knowledge that AI agents need to run in production.

Jedify sells software that automatically assembles customer-specific “context graphs” on top of a company’s existing data and knowledge systems. The platform connects a company’s operational data stored in warehouses, customer relationship management, and financial systems with unstructured materials scattered around it, such as documents, playbooks, Slack threads, and meeting notes. From there, Jedify says, it builds a semantic model that keeps up-to-date on how a business defines metrics, how records are related, and who can see what.

Enterprise AI adoption has suffered from precisely this gap. Large-scale language models can produce fluent answers, but unless that context is provided at runtime, they cannot reliably determine which revenue definitions apply, which customer records are current, or which operational assumptions are important. Without this, the company claims, agents would hallucinate or burn tokens processing irrelevant information.

“For agent workflows to really work in a large enterprise, it requires a very deep understanding of the business,” said co-founder and CEO Assaf Henkin. “Enterprise data is fragmented across systems, definitions, permissions, and workflows. Jedify transforms that fragmented knowledge into a live context graph that enables agents to generate accurate, cost-effective, business-ready answers.”

Jedify positions itself as a model-agnostic layer separate from model providers. The company argues that companies that pass data to the same vendors that sell tokens face disproportionate incentives, citing recent moves by OpenAI Group PBC, Anthropic PBC, and Google LLC to provide forward-deployed engineers and professional services to customers. Jedify says an independent context layer avoids that conflict and lock-in of a single vendor, which conflicts with the governance requirements of most large organizations.

The technology behind this platform, which Jedify calls Semantic Fusion, is patent pending. The company says that with each interaction, the customer’s context graph becomes more accurate, turning it into a unique asset that grows over time. Jedify is also working with Snowflake Inc. on integration with Cortex AI tools such as Semantic Views and Cortex Analyst.

The funds will be used for product development, market development and hiring.

Norwest led the Series A round with strategic investment from Snowflake Ventures and participation from existing backers S Capital VC and Cerca Partners, and new investor Oceans Ventures. Assaf Harel, partner at Norwest, will join Jedify’s board of directors.

“Jedify solves a fundamental problem by autonomously blending structured and unstructured data to create a context graph that gets smarter with each interaction,” said Harrell. “Its value compounding and model-agnostic approach gives businesses flexibility rather than lock-in, which is exactly the durable infrastructure layer we are looking for.”

Jedify raised a $8.5 million seed round led by S Capital VC in September 2023, bringing total funding to more than $33 million.

Photo: Jediify