By Asep Darmawan 
For the past two years, the corporate narrative surrounding artificial intelligence has been defined by a singular, persistent myth: the "turnkey" solution. AI vendors have promised that their agents could be deployed like off-the-shelf software, instantly transforming business operations with a few clicks. Yet, the reality has been far more sobering. Most enterprise AI implementations fail to gain traction because they lack an understanding of the business’s unique internal language—how a specific firm defines "revenue," which employees are authorized to access sensitive documents, or the nuanced operational assumptions that govern daily workflows.
Jedify, a New York-based startup, is positioning itself as the bridge across this chasm. Today, the company announced it has secured $24 million in a Series A funding round led by Norwest Venture Partners. The raise, which brings Jedify’s total funding to approximately $33 million, underscores a growing industry realization: in the age of generative AI, context is more valuable than the model itself.
The Core Problem: Why AI Agents Struggle in the Enterprise
The "AI agent" is meant to be an autonomous worker capable of executing tasks, but an agent is only as good as the information it can access. Currently, most enterprise AI tools are "blind" to the complex web of relationships that define a company’s operations.
When a standard AI model is pointed at a company’s data, it often lacks the ability to navigate the silos where that information lives. Whether it’s data warehouses, SaaS applications like Slack or Notion, or unstructured meeting transcripts, most AI agents cannot synthesize these disparate sources into a coherent picture.
"Unless you put in the effort to train a model on the specifics of your business, it’s unlikely to understand how your company operates," says Assaf Henkin, co-founder and CEO of Jedify. "AI vendors promote their products as turnkey solutions, but the reality is that the gap between a generic model and a business-ready agent is enormous."
This gap is precisely why many AI companies have been forced to deploy armies of integration engineers to help customers manually tether their systems to new AI tools—a process that is neither scalable nor cost-effective.
How Jedify Builds the "Context Graph"
Jedify’s solution is a platform that connects to an enterprise’s knowledge sources via APIs to build what the company calls a "Context Graph." Unlike traditional metadata catalogs or basic knowledge graphs, Jedify’s architecture is multi-dimensional. It captures the complex interplay between entities, data, human roles, permissions, and institutional workflows.
The platform pulls from a vast array of sources:
- Structured Data: Databases, data warehouses (like Snowflake), and BI tools.
- Unstructured Data: Technical documentation, internal reports, codebases, and communications.
- Real-time Interaction: Slack channels, meeting recordings, and CRM data.
By aggregating this data, the Context Graph allows an AI agent to narrow its focus. Instead of searching through the entire corporate digital footprint for every query, the agent uses the graph to understand which information is relevant to the specific task at hand.
Chronology and Strategic Backing
The journey for Jedify has been marked by rapid validation from both the venture capital community and the data industry.
- Early Development: Founded to address the fragmentation of enterprise data, the company focused on building a model-agnostic layer that could sit atop existing infrastructure.
- Strategic Partnerships: The recent Series A, led by Norwest, included participation from existing backers S Capital VC and Cerca Partners, as well as new investor Oceans Ventures.
- The Snowflake Integration: Perhaps the most significant validation comes from Snowflake, which participated as a strategic investor. Snowflake is currently integrating Jedify’s technology into its own AI product suite, including its Cortex AI service, Semantic Views, and CoWork tools. This partnership signals that major data platforms recognize that they cannot solve the "context" problem alone.
Real-World Application: The Case of Kiteworks
To understand the utility of the Context Graph, one need only look at Kiteworks, a compliance and security company. Kiteworks utilized Jedify to connect its disparate stack—including Snowflake, Tableau, Notion, and internal corporate playbooks—into a unified, agentic interface.
According to CEO Assaf Henkin, Kiteworks wanted to equip its sales and account management teams with a tool that functioned as both a dynamic dashboard and a real-time conversational assistant.
"When they go into a customer conversation, Jedify builds for them, on the fly, everything they need to know," Henkin explains. "During the conversation, they can get very specific details surfaced proactively." By mapping the relationship between a client’s account status, technical compliance requirements, and recent internal discussions, the Jedify-powered agent provides insights that would take a human analyst hours to retrieve.
Addressing the "Permissions" Hurdle
One of the most daunting challenges in enterprise AI is governance. If an AI agent has access to all company data, how do you prevent an intern from accessing a CFO’s revenue projections?
Jedify addresses this by inheriting existing permissions from the customer’s identity systems, file systems, and database access rules. The platform supports granular row-, column-, and table-level access controls. Furthermore, it provides observability and governance tools that allow administrators to monitor what their AI agents are "seeing" and doing. By creating a layer that respects existing security infrastructure, Jedify removes a primary roadblock to AI adoption: the fear of data leakage or unauthorized access.
Implications: The Shift Toward Proprietary Context
As AI models become more commoditized and interchangeable, the industry is shifting its focus from the model to the data that feeds it.
Henkin argues that for companies attempting to build this context layer on their own, the costs are prohibitive. As corporations face increasing pressure to manage "token usage" and AI-related infrastructure costs, relying on brute-force training is no longer sustainable. Jedify offers a more efficient alternative, effectively serving as a "brains" layer for any LLM.
"The big thing is that not all of your data is in [the big cloud providers]," says Henkin. "Most of your knowledge is not there, so it’s a bit of a disadvantage that they actually have. Jedify is complementary because it bridges those gaps."
Future Outlook
With $24 million in fresh capital, Jedify plans to scale its product development and expand its go-to-market efforts. The company is currently targeting mid-market and large enterprises with mature data stacks—sectors such as gaming, industrials, and consumer packaged goods are already showing significant interest.
The company’s vision is built on a bold, long-term bet: that as AI models become more powerful, the primary differentiator for a successful business will not be the model itself, but the proprietary "context" that makes that model uniquely effective for their specific operations.
In a world where every enterprise is racing to implement AI, Jedify is betting that the winners won’t be those with the smartest models, but those with the most context-aware agents. For now, the startup is well-positioned to prove that in the complex, siloed world of corporate data, the most valuable asset isn’t the code—it’s the map.
rifanmuazin 
Azzam Bilal Chamdy