Logo
100% Serverless • Enterprise-Grade • AI-Ready

Platform Architecture

Modern, Scalable, and Built for the Future

A fully serverless, AI-ready clinical trial management platform built on proven technology with enterprise-grade security and compliance.

Complete Platform Architecture
Analytics & Semantic Layer Architecture

Platform Architecture

A fully serverless, AI-ready clinical trial management platform built on proven technology with enterprise-grade security, compliance, and modern analytics — designed for scalability and seamless integration.

Zero maintenance overhead with fully managed services. Built on AWS cloud infrastructure with automatic scaling. Start small and scale seamlessly as your trials grow without infrastructure management.

  • Zero maintenance overhead with fully managed services
  • Built on AWS cloud infrastructure with automatic scaling
  • Start small and scale seamlessly as your trials grow
  • No infrastructure management required
  • Dedicated single-tenant stack per client, in your chosen AWS region
  • Uptime SLA available with managed plans

The Third Data Layer

Transactional systems answer what is happening now. Analytical systems answer what happened at scale. The temporal knowledge graph answers what is connected, how it evolved, and why.

Transactional

"What is happening now?"

The operational databases the trial runs on — visits, forms, consents, queries — optimized for the current state of every record.

  • Current subject status
  • Open queries and tasks
  • Live operational lookups

Analytical

"What happened at scale?"

The warehouse and governed semantic layer — aggregates, trends, and KPIs computed from centrally defined metrics.

  • Enrollment rates and trends
  • Governed KPIs
  • Cross-study aggregates

Temporal Knowledge Graph

"What is connected, how did it evolve, and why?"

The third layer: entities and relationships with time on every edge — the connective tissue the first two layers cannot express.

  • Amendment provenance
  • Affected sites and activities
  • Before/after history of any change

A Living Graph from the Audit Trail

Business events stream from the clinical system into Graphiti, which extracts and resolves entities and relationships over time into a FalkorDB temporal knowledge graph — served to AI assistants through MCP.

Clinical System

CTMS / EDC transactions

Business Events

Stream / CDC from the audit trail

Graphiti

Entity & relationship extraction, resolution, temporal processing

FalkorDB

Temporal knowledge graph, queried in Cypher

MCP Services

Fixed-shape graph tools over MCP

AI Assistant

Answers with provenance

Every edge in the graph carries valid-from / valid-to — the graph remembers what was true, when.

The auditor's question

“Explain everything around Amendment 4 — who requested it, who approved it, which sites were affected, what happened after.”

Designed to be answered in one graph traversal — instead of days of joining audit tables.

A Two-Tier Ontology — Designed, Not Inferred

The graph is built on a deliberately designed clinical ontology: first-class trial entities in Tier 1, and process entities the graph layer promotes in Tier 2.

Tier 1

First-class trial entities

StudySiteSubjectPatientInvestigatorStudy PersonnelStudy Event / VisitConsentAdverse EventCRF FormUser

Tier 2

Promoted by the graph layer

ProtocolProtocol AmendmentChange RequestReviewApprovalDeviation

CDISC SDTM/ODM exports are modeled as export-artifact nodes with derived-from provenance edges — records of what was exported and where it came from, not business entities.

“AI can help populate and evolve the graph. The enterprise owns the ontology.”

The governance principle behind the knowledge engine.

Reference Architecture

Two parallel paths from the clinical applications — governed numbers through the analytical path, relationships and time through the graph path — converging at the MCP layer that serves the assistant.

Clinical Applications

The systems the trial already runs on

Analytical Path — Governed Numbers

Transactional DB

Operational records

Warehouse

Modeled, historized data

Semantic Layer

Centrally governed KPIs

Graph Path — Relationships + Time

Business Events

Stream / CDC

Graphiti

Extraction, resolution, temporal processing

FalkorDB

Temporal knowledge graph

MCP Layer

Both paths served through one protocol

AI Assistant

Agents run on Amazon Bedrock (AWS)

Enterprise-Grade Capabilities

Built on modern cloud infrastructure with security, compliance, and scalability at its core

100% Serverless

Zero maintenance overhead with fully managed services. Start small and scale seamlessly as your trials grow.

Rock-Solid Foundation

Built on Frappe and ERPNext—the same platform powering Zerodha, India's leading fintech company.

Healthcare-Grade Security

SSO/OpenID Connect authentication, fine-grained role-based access control, and audit logging.

Optimized Performance

PostgreSQL, optional Redis caching, and S3-compatible storage for lightning-fast operations.

Multi-Platform UI

React/Next.js web app, native iOS/Android apps, and mobile-optimized interfaces.

Built-In Lakehouse

Modern dbt-powered analytics with PostgreSQL. Supports Snowflake and Databricks.

API-First Design

100+ documented REST APIs. Use as headless CTMS with standards-based integration.

AI-Ready Platform

Includes MCP server for building agentic AI applications and intelligent workflows.

Dedicated Deployment

A dedicated single-tenant stack per client, deployed in the client's chosen AWS region.

Clean Architecture

Separation of concerns with well-defined layers for maintainability and extensibility.

Standards-Based

FHIR-ready, HL7 compatible, and follows industry best practices for interoperability.

Rapid Deployment

Cloud-native design enables quick deployment and instant scaling for your trials.

API-First Architecture

Comprehensive REST APIs with detailed documentation for seamless integration and extension

Complete API Documentation

Interactive API reference with code examples in multiple languages

100+ REST Endpoints

Comprehensive coverage of all platform features and data access

Secure & Authenticated

OAuth 2.0, API keys, and role-based access control

Headless CTMS Mode

Use Zynomi as a backend for your custom applications

Ready to Experience Modern CTMS?

Discover how Zynomi's architecture delivers security, scalability, and speed for your clinical trials.