Enterprise artificial intelligence context startup Jedify Inc. today formally launched Data Agents, a product that lets business teams set up autonomous AI analysts by describing the work they want done in plain language.

Data Agents takes on a longstanding limit of business intelligence software, where analysts have to build dashboards ahead of time around the questions they expect people to ask. Generic AI assistants have tried to fill that gap. In Jedify’s telling, those tools end up guessing at what a business means from its raw tables and schemas.

The company’s core product is a context graph that pulls structured data and documents into a single model of how a business works, down to metric definitions and record-level permissions. Data Agents runs on top of it. Because joins and schema mapping are already worked out inside the graph, building an agent comes down to describing the question.

Users decide when an agent runs, whether on request, on a schedule or when a condition they have defined is met. A market that starts to underperform is the kind of anomaly an agent is meant to catch on its own, and Jedify’s deep research engine then works out what most likely drove the change.

Data access is tied to the person asking the question. Anything that user is barred from seeing, even a single row in a table, stays out of the agent’s results. On cost, Jedify claims its agents burn far fewer tokens per question than ones that have to explore raw schema at query time.

Teams can also work with the agents outside Jedify’s own interface. Under the Model Context Protocol, each one appears as a tool that can be called from Claude, ChatGPT, Slack or any agent platform a company runs. Results can go straight to email or Slack too.

Jedify is pitching the product at data teams and the business staff they support, arguing that analysts who keep fielding the same dashboard request can hand it off to an agent. On the business side, people in finance, revenue operations and sales can set up their own agents.

Co-founder and Chief Technology Officer Adi Elimelech said enterprises are moving analytics “from answering questions to doing meaningful work.” The goal, he said, is for teams to build “an analytical workforce” matched to how their business operates and reach those agents from the tools employees already use.

Online casino game developer RubyPlay Ltd. is already using the product. Its team set an agent up once and scheduled it to run every day, and the report now arrives without anyone asking for it.

“It even flagged us on its own that not all of our target markets were covered yet, which told us it was reasoning through the task rather than just running a script,” said Avi Baruch, vice president of data and business applications at RubyPlay.

Data Agents is available now through Jedify’s Agents Hub, and companies can start on it for free. Jedify is a venture capital-backed startup that has raised more than $33 million, including a $24 million Series A round led by Norwest in June.

Image: Jedify