Jedify launched Data Agents on October 28, 2026, a product that lets data and business teams build autonomous analysts in plain language, without writing code or building workflows. The agents monitor business metrics, investigate changes and run recurring analyses, drawing on Jedify’s context graph to understand an organisation’s specific metrics, relationships and business rules, rather than inferring meaning from raw database tables.
From Dashboards To Analysts That Come To You
Traditional business intelligence, Jedify argues, asks people to define their questions in advance and build a dashboard to answer them. If the question changes, the dashboard has to change too. Generic AI assistants have the opposite problem: they can answer a question, but they start from raw tables and schema, so they may misunderstand what a metric means in that particular company, or pull the wrong data entirely.
Data Agents are meant to sit between those two approaches. A user describes the analytical outcome they want in plain language, and Jedify creates and refines the agent conversationally. The agent then runs on demand, on a schedule, or when a defined condition is met, so a recurring analysis happens without someone asking for it each time. When a metric moves, the agent investigates why, rather than simply reporting that it moved.
Jedify’s chief technology officer and co-founder, Adi Elimelech, framed the shift as a move from answering questions to doing meaningful work. Enterprises, he said, want an analytical workforce tailored to how their business operates, available to employees wherever they already work, not another dashboard that nobody opens.
What The Agents Actually Do
The core promise is root-cause investigation. Grounded in Jedify’s context graph and deep research engine, Data Agents monitor metrics, investigate anomalies and identify likely drivers behind a change. Instead of waiting to be asked the right question, they surface what needs attention — a gap, an anomaly, an underperforming market — the moment it emerges, Jedify says.
Two design choices are meant to make that safe at enterprise scale. First, agents are permission-aware by default: they inherit the requesting user’s existing data permissions, so an agent never surfaces a row, column or metric that the user is not already entitled to see. Second, because agents reason from Jedify’s pre-resolved context graph instead of exploring raw schema at query time, they use far fewer tokens per question than a schema-inferring agent, according to the company.
Data Agents are available now through Jedify’s Agents Hub. Through the Model Context Protocol, an agent can also be exposed as a tool that teams call from Claude, ChatGPT, Slack or any agent platform they run, with results delivered directly through Slack and email as well.
Early Customers And The Test Ahead
Jedify pointed to two early users to illustrate the product in practice. Avi Baruch, vice president of data and business applications at RubyPlay, said his team set up an agent once, scheduled it daily, and now receives the report rather than chasing it, adding that the agent flagged on its own that not all target markets were covered, which suggested it was reasoning through the task rather than running a script. A vice president of enterprise technology at ESL FACEIT Group said that when an agent flags a specific cost line moving 10 percent in a month, rather than burying it in a trend chart, it gives the team something it can actually act on.
Those are company-supplied testimonials, and they should be read as such. The real test for Data Agents will be whether they hold up outside a friendly pilot: whether their explanations stay accurate when data definitions change, whether permission inheritance works cleanly across messy real-world access controls, and whether teams trust an agent’s root-cause claim enough to act on it without re-running the analysis by hand.
Jedify is also using the launch to reach new buyers. It will present Data Agents at the Gartner Supply Chain Planning Summit, and it is holding a webinar on October 29 with a live product overview and a customer session with ESL FACEIT Group, aimed at revenue teams using the product to automate analysis.
Why It Matters
This matters because analytics teams are drowning in dashboard requests, while business teams still wait days for a simple answer about why a number moved. If plain-language agents can genuinely investigate, rather than just retrieve, they could shift analytics from a reporting function to a monitoring one, where the first alert already carries a plausible explanation. The same shift is visible in agentic platforms for the whole workforce, accountants working with AI on one platform and AI handling live customer conversations: the winners will be the tools that act inside real workflows, not the ones that only answer questions.
The risk is over-trust. An agent that confidently explains a metric movement with the wrong driver is worse than a dashboard that says nothing, because it sends a team down the wrong path with confidence. Jedify’s context graph is its answer to that risk, but customers will judge it on the accuracy of its explanations, not on the elegance of its architecture.
Background
Jedify, which describes itself as an autonomous context graph for data-intensive agentic applications, launched Data Agents in New York on October 28, 2026, available now through its Agents Hub, with a customer webinar scheduled for October 29, 2026.
Source: Jedify announcement
0 Comments