Jedify is betting that enterprise AI fails not because the models are wrong, but because the business context feeding them is a mess. The Israeli startup raised $24 million in a Series A led by Norwest, with Snowflake Ventures taking a strategic position, bringing total funding to $33 million. The company builds context graphs for enterprise AI agents<\/a>, pulling structured data from warehouses and CRMs together with unstructured sources like Slack logs and meeting recordings into a continuously updated semantic model agents can query at runtime.<\/p>\n The organization that feels this story most is one already running agentic workflows and watching them hallucinate or burn compute on irrelevant data pulls. Jedify’s core argument, that fragmented enterprise data is the actual chokepoint in production AI, is one that CDOs have heard from their own teams for two years. The question isn’t whether the problem is real. It’s whether a purpose-built context graph layer is the right fix, or whether a well-governed semantic layer inside an existing data platform gets you 80% of the way there for less integration overhead.<\/p>\n The Snowflake angle deserves more scrutiny than the press framing gives it. Snowflake Ventures investing while Jedify integrates with Cortex AI products including Semantic Views and Cortex Analyst isn’t a neutral partnership. It positions Jedify as a complement to Snowflake’s own semantic ambitions, which means Jedify’s model-agnostic pitch sits inside a distribution relationship that quietly favors one data platform. That’s not disqualifying, but CDOs evaluating Jedify should pressure-test whether deep Snowflake integration actually delivers the vendor-independence the company is selling.<\/p>\n The broader pattern here is that the race to solve enterprise AI context is splitting into two camps: platform vendors extending their existing semantic layers upward into agents (Snowflake, Databricks, Microsoft Fabric), and independent players like Jedify building a context graph that sits above all of them. Independents win this fight only if enterprises genuinely run heterogeneous data stacks where no single platform can claim the full semantic picture. If your data estate is consolidating onto one cloud data platform, Jedify’s value proposition shrinks with every migration you complete.<\/p>\n A context graph is a structured map of an enterprise’s business knowledge, capturing not just data but the relationships between it: what a “customer” means in the sales system versus the finance system, which metrics are authoritative, who has permission to see what, and how domain-specific terms translate across departments. Think of it as the organizational dictionary an AI agent consults before answering, rather than guessing. Without it, agents produce answers that are technically coherent but business-wrong.<\/p>\nWhat this means for your business<\/h2>\n
Concept deep-dive: Context graph<\/h2>\n