Clinical study information was distributed across multiple applications, including regulatory document management systems and clinical trial management system platforms. With no common identifier linking records across these systems, consolidating and reconciling clinical data was a complex process.

The client needed to bring structured and unstructured data together at scale while maintaining the traceability and governance required for clinical research.

Manual reconciliation across systems was also time-consuming and prone to errors. As a result, study-ready data could take several days to become available, limiting the speed at which teams could access and act on critical information.

• Iris designed and implemented a Databricks-based data platform on AWS to integrate clinical data from multiple systems into a centralized and governed environment.
• A metadata-driven ingestion framework using a Bronze-Silver-Gold architecture was established to streamline the onboarding of new data sources while maintaining consistent data lineage.
• Iris also built a reconciliation and matching engine that generated ranked, explainable candidate matches across disparate systems. Human-in-the-loop workflows enabled users to review, validate, and approve matched records, providing an additional layer of control and accuracy.
• Enabled secure sharing of approved study, site, and regulatory data through Databricks and Delta Sharing.

The solution helped the client establish a more connected and responsive clinical data environment. Key outcomes included:

• Created a single trusted view of clinical data across multiple systems.
• Reduced data availability time from 3–5 days to near real-time, improving operational responsiveness.
• Reduced manual reconciliation efforts through automated, auditable matching and approval workflows.
• Approved study, site, and regulatory data was then made available to client environments through Delta Sharing, enabling secure, zero-copy data exchange.

Clinical trial information was distributed across multiple systems and data sources, making it difficult to integrate, interpret, and use the information effectively.

The data included unstructured content such as protocols, administrative correspondence, schedules, and supporting documents, along with structured data from IRB systems, CTMS platforms, and operational databases.

This fragmented data landscape limited the ability to generate timely insights, streamline workflows, and scale AI-driven decision-making across clinical research operations.

• Iris analyzed the requirements and developed a scalable intelligence platform to bring disparate clinical trial data together and transform it into trusted, actionable information.
• The solution included robust data ingestion pipelines capable of processing more than 80,000 protocol documents at scale and converting unstructured content into high-quality structured data. This established a reliable data foundation for AI/ML models and supported the expansion of the platform's knowledge base.
• We also built human-in-the-loop (HITL) validation workflows to enable human review and oversight of AI-generated outputs. The workflows helped refine results while strengthening governance, compliance, and quality controls required for clinical research.

The platform created a reliable data backbone for clinical research and helped the client make better use of its growing data landscape. Key outcomes included:

• Accelerated AI-driven insights through a reliable and structured data foundation
• Reduced manual effort involved in data preparation
• Faster onboarding of new data sources
• Improved accuracy and trust in AI-generated insights
• Strengthened compliance and quality controls for clinical research use cases

The client's data platform consisted of numerous ETL pipelines, database repositories, and codebases, making it difficult for developers to understand dependencies and trace data flows across the environment.

Developers spent significant time investigating failures, tracing dependencies, understanding the potential impact of changes, and working through change requests. This was further complicated by business knowledge, technical documentation, and code repositories being maintained in separate silos.

The lack of a unified view made impact analysis and troubleshooting more difficult and increased the time required to resolve tickets and implement changes.

• Iris built a unified knowledge platform that brought together code lineage, dependency mapping, and business documentation to give developers relevant context in one place.
• The platform used Graphify to map code lineage and dependencies and Confluence to bring in business and technical documentation. Developers could use Jira tickets as inputs to retrieve relevant context, understand potential impact, and access recommended resolutions from these knowledge sources.
• By surfacing dependencies, data lineage, and implementation guidance, the solution reduced the need for manual investigation. Context-aware engineering workflows also helped developers move more quickly from identifying an issue to determining the appropriate resolution.

The solution helped streamline engineering workflows and reduce the time spent investigating and resolving development issues. Key outcomes included:

• Reduced ticket resolution and change implementation timelines from days to hours
• Improved developer productivity through faster access to dependency and data lineage information
• Simplified impact analysis and troubleshooting by bringing technical and business knowledge together
• Accelerated bug resolution and change delivery through context-aware engineering workflows.

The client's Clinical Trial Management System (CTMS) supports clinical studies across the research lifecycle, from study setup through closeout. The platform also helps streamline clinical trial operations, improve billing compliance, and provide actionable reporting and analytics.

As the product continued to evolve, the client needed to enhance its core capabilities while maintaining the scalability, performance, security, and reliability required to support growing clinical research needs.

The platform also required modernization of its integrations and continued development of business-critical capabilities to improve interoperability and support future growth.

• Iris acted as a strategic product engineering partner, supporting the evolution of the CTMS platform and its long-term product roadmap.
• The team owned end-to-end delivery of business-critical features, covering requirements analysis, solution design, development, testing, and release. Iris also drove continuous product and technology improvements to enhance platform scalability, performance, security, and maintainability.
• As part of the modernization efforts, Iris enhanced APIs and integrations to improve interoperability and prepare the platform for future growth. The team worked closely with product management and business stakeholders to translate customer needs into relevant product capabilities.
• Iris contributed across key functional areas, including Protocol Creation, Clinical Trial Calendars, and Payments & Financial Management.

The product engineering engagement helped strengthen the CTMS platform across accessibility, integration, security, and reliability. Key outcomes included:

• Achieved WCAG 2.0 compliance for the client's research portal, improving accessibility and user experience
• Modernized payment integrations by migrating from SOAP to REST APIs, improving interoperability and scalability
• Strengthened product security and reliability through proactive vulnerability remediation and critical security enhancements