AI agents turns business intent into inspectable, production-grade visual data workflows—running natively on Databricks, Snowflake, and BigQuery
Prophecy, the AI data prep & analysis platform, today announced Prophecy v4, bringing AI agents based on Claude Code to business users for visual data preparation and analysis. Prophecy v4 generates visual data workflows that users can visually inspect, refine, and trust—while producing production-grade code stored in Git and executing with governance and scale on customers’ cloud data platforms.
As AI begins to generate more of the data logic, the bottleneck shifts from generation to validation and trust. Prophecy v4 addresses this by making AI-generated logic visual and reviewable, avoiding the slow, error-prone process of validating pages of AI-written SQL or Spark code. Instead, users collaborate with specialized agents that generate workflows, compile and run them against the target platform, then lift the resulting logic into visual components that remain synchronized with underlying code.
“Coding without AI has become unthinkable in the past two years, we are bringing the same power to millions of business data users” said Raj Bains, CEO at Prophecy. “For our customers, results that took days or weeks are being done in minutes”
Why this matters: data prep is still the analytics bottleneck
Most business data users aren’t programmers. They often start in spreadsheets, but as data becomes more complex—multiple sources, joins, and quality rules—work gets pushed into ticket queues for central data teams. The result is slow iteration, growing backlogs, and lost momentum. Legacy desktop tools made prep more accessible, but they were built before cloud platforms, Git-based collaboration, and AI-driven workflows.
Prophecy v4 is designed to replace legacy desktop tools and deliver a step-change in speed, quality, and scale by combining agent-driven productivity with cloud-native execution.
Key capabilities in Prophecy v4
1) AI radically increases productivity
Prophecy v4 brings an agent-driven, human-in-the-loop workflow to data prep & analysis. Users describe business intent, agents generate workflows as code, validate results by compiling and running them, and present the logic visually so users can inspect and refine quickly—keeping visuals and code in sync.
2) AI collapses the data tool stack
Working with AI is iterative and fluid: asking questions, inspecting data, refining logic, getting insights, and operationalizing final results. Prophecy v4 unifies these steps into a single canvas, creating vibe data analysis where users are able to iteratively get to answers through a conversational interface with an agent, and results can be trusted by visually inspecting the workflow.
3) Built for trust, governance, scale, and production
Prophecy v4 is designed for enterprise realities:
- Native cloud execution on Databricks, Snowflake, and BigQuery
- Open and trusted: workflows stored as native code in Git
- Governed: identity, access, and policies inherited from the data platform
- Auditable and reproducible by default
Because workflows are production-grade from day one, business-built logic can move cleanly to platform teams without rewrites.
The result is hours of manual work reduced to minutes of review.