CData Software, a leading provider of data connectivity solutions, announced the launch of three new developer-focused products designed to eliminate data integration bottlenecks for engineers building artificial intelligence applications: Connect AI Developer Edition, the CData Connect AI Python SDK, and the CData CLI.
The coordinated product rollout represents a significant shift in how developers interact with enterprise data layers. While AI projects frequently stall because navigating rigid IT security protocols can delay accessing live production data, CData’s new ecosystem allows developers to securely tap into hundreds of enterprise platforms including Salesforce, Snowflake, NetSuite, Microsoft 365, and Workday—using standard SQL, Python workflows, the command line, and the Model Context Protocol (MCP).
“Developers have been forced to choose between moving fast and meeting the governance their company requires. That tradeoff doesn’t hold up anymore,” said Raviv Levi, Chief Product and Technology Officer at CData. “The same live, governed access that IT trusts and the business depends on should follow developers into the terminal, the Python environment, and the IDE. That’s what these releases do.”
Unlocking Managed Data Connectivity with Three New Developer Tools
CData’s new offerings directly target the data layer friction points that frequently cause enterprise AI initiatives to stumble, such as API rate limits, authentication drift, and schema changes. The launch introduces three specific developer entry points:
Connect AI Developer Edition (Free): This tier gives developers full, free access to CData’s enterprise feature set. It automatically exposes legacy APIs as standard, queryable data layers while natively managing token refreshes, pagination, and API throttling behind the scenes. It features complete MCP server support to interface out of the box with popular AI development environments and coding assistants like Claude Code, Codex, Cursor, and LangChain.
CData Connect AI Python SDK (Open Source): Built on the familiar DB-API 2.0 specification, this open-source, MIT-licensed SDK allows developers to pull live, governed corporate metrics directly into native Python workflows. Engineers can utilize familiar data-science packages like pandas and SQLAlchemy to communicate with enterprise systems without learning custom APIs.
CData CLI (Command-Line Interface): A terminal-native tool designed so coding assistants can programmatically query and scaffold connectivity without needing manual documentation review. The initial release provides robust support for CData’s JDBC and Python drivers, with ADO.NET and ODBC capabilities planned for subsequent updates.
Also Read: Kurrent Launches Capacitor to Drive Human-Agent Code Collaboration
Mitigating Token Waste and Schema Drift in Agentic AI
A key technical challenge in multi-agent configurations is managing unexpected schema drift. For example, if an operations manager creates a custom field within a CRM on Monday night, an unguided AI agent executing an automated report on Tuesday morning may generate inaccurate or corrupt data outputs due to missing context. Connect AI resolves this issue through real-time metadata discovery, pinging the target database on every query to ensure the AI model is always acting on the platform’s current state.
Furthermore, to combat the problem of “over-permissioning” in agentic setups, the software introduces Toolkits. This feature allows engineering teams to package precise data scopes into unique MCP Server URLs tailored for specific tasks. This ensures that an active AI worker only receives the exact data parameters required for its given objective, keeping security tight and preventing unexpected token drain.
“The need to accelerate development exists everywhere,” said Jerod Johnson, Director of Technology Evangelism at CData. “Product owners are asking, ‘You have an AI coding assistant, why isn’t this shipping faster?’ And people are still running into the problem — I can ship that feature, but I also have to build the connector that gets the data to it.”
All three developer-centric solutions are fully live and available for download. Engineering leads, enterprise data architects, and software developers can review setup protocols, download the open-source Python driver packages, and join the developer community by visiting the official CData Developer Center.