Congrats to Casepoint for the announcement that Casepoint expands Casepoint IQ with its first purpose-built AI agents! Here’s a portion of the press release, the full release is here!

Casepoint Expands Casepoint IQ With First Purpose-Built AI Agents

Relevance Determination Agent and Issue Coding Agent are part of a growing portfolio designed to automate complex government, legal, and compliance workflows with human oversight built in.

WASHINGTON, September 23, 2026 – Casepoint, a leading provider of secure, unified government, legal, and compliance solutions, today announced a portfolio of purpose-built agents powered by Casepoint IQ™, including the Relevance Determination Agent and the Issue Coding Agent. Now live within Casepoint’s eDiscovery application and in use by early access customers, the agents mark the first deployment in a broader pipeline of specialized AI agents Casepoint is building for government, legal, and compliance teams. 

The announcement follows the launch of Casepoint IQ earlier this month, the intelligence foundation spanning Casepoint’s agentic AI, assistive AI, machine learning and predictive coding, and MCP enablement. Rather than building standalone AI tools for individual tasks, Casepoint is developing a growing portfolio of agents that can operate within the same secure platform, data environment, and governance framework customers already use. 

“Government and legal teams don’t need one general-purpose AI agent trying to do everything. They need specialized agents that understand the work, operate within established controls, and can ultimately work together across complex workflows,” said Paul Colangelo, CEO at Casepoint. “That’s what we’re building on Casepoint IQ. The Relevance Determination Agent and the Issue Coding Agent are the first proof points, with a much broader pipeline of agents being developed across the workflows our customers manage every day.” 

Putting Agents to Work 

Casepoint IQ agents target document-intensive work that traditionally requires significant human review: 

• Relevance Determination Agent evaluates documents to determine whether they are relevant or responsive, provides the reasoning behind each determination, and assigns a likelihood score. Every human and AI decision is recorded in an audit trail. 
• Issue Coding Agent evaluates documents against specific legal, factual, or investigative issues defined for a matter. An optional post-run quality-control process calculates final recall and precision accuracy at the individual issue-code level. 

Both agents use a proprietary multi-model process that analyzes and cross-checks outputs before returning results to a reviewer. Built-in workflows require users to validate performance against a sample set before the agents run across the full scope of documents. Recall and precision metrics measure accuracy, while estimated time and cost savings give legal teams visibility into the potential impact before proceeding. The approach combines agentic automation with measurable performance and human oversight. 

“These agents aren’t simply generating an answer and asking users to trust it,” said Vishal Rajpara, Founder and Chief Technology Officer at Casepoint. “They show their reasoning, measure their performance, and pause for human validation at critical points. That architecture gives us a repeatable foundation for building increasingly sophisticated agents while maintaining the transparency and control that legal and government work requires.” 

Building a Portfolio of Specialized Agents 

Casepoint is actively developing additional agents designed to address other labor-intensive legal, investigative, and information-governance tasks. Casepoint’s architecture is designed to allow specialized agents to work sequentially across a workflow to move a matter from relevance analysis to issue identification and other review and data-handling tasks without requiring teams to move between separate technologies. 

The strategy reflects Casepoint’s broader vision for Casepoint IQ: agentic AI as an integrated operating layer across its unified platform rather than a collection of disconnected AI applications. 

Both currently available agents run within Casepoint’s unified platform security boundary and operate under existing customer permissions, governance, and access controls. The Relevance Determination Agent and Issue Coding Agent are available for demonstration alongside the broader Casepoint IQ platform. To request a demonstration, visit casepoint.com/iq-agents.

Congrats again to Casepoint for the announcement that Casepoint expands Casepoint IQ with its first purpose-built AI agents! Important functionality!

So, what do you think? Please share any comments you might have or if you’d like to know more about a particular topic.

Disclosure: Casepoint is an Educational Partner and sponsor of eDiscovery Today

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by my employer, my partners or my clients. eDiscovery Today is made available solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscovery Today should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.