The competition, which drew 50 submissions from 30 organizations, tasked participants with applying machine learning to automate clinical outputs while maintaining an auditable trail from initial study objectives to final reporting. TLFGenix operates as an end-to-end pipeline that ingests Statistical Analysis Plans and ADaM datasets to produce submission-ready documents. The system uses six specialized agents to digitize analysis plans, generate driver scripts, and verify consistency within a secure, network-isolated environment.

Deepak Ananthan, head of clinical services and AI scaling at Zifo, noted that the platform prioritizes human oversight by requiring expert sign-off at critical junctures. Rather than replacing statistical judgment, the software uses a lineage graph to connect endpoints and results, allowing for rapid regeneration when specifications change. This approach addresses the industry-wide need for greater transparency and compliance in regulated trial environments. Zifo plans to present the technology at the 2026 CDISC US Interchange in Denver this October, further aligning its work with the CDISC 360i initiative for connected, end-to-end research data flows.