Data Architect

Job Description:

We are seeking a Senior Data & Integration Architect to lead data model design and downstream integration strategy for a large-scale policy administration system modernization. You will reverse engineer legacy mainframe data structures, forward engineer them into modern SQL/MongoDB document schemas, and define how data flows to downstream consumers. You will leverage purpose-built AI agents to accelerate reverse engineering, model generation, and documentation bringing human judgment to validate and refine AI-produced outputs.

Key Responsibilities:

• Design and maintain technology-agnostic Logical Data Models (entities, relationships, cardinality, PK/FK)
• Transform LDMs into modern physical schemas applying aggregate-oriented and DDD patterns
• Reverse engineer IMS hierarchical segments and DB2 tables extract business entities from physical storage structures without original design documentation
• Interpret COBOL copybooks as data structure definitions and map legacy field types to modern equivalents
• Define embedding vs. referencing strategies, versioning patterns, and collection boundaries for target database platforms
• Design downstream integration patterns REST APIs, event streaming (Kafka/MQ), Change Data Capture (CDC), and data distribution to consuming systems
• Direct and validate AI agent pipelines for automated reverse engineering, ERD generation, data dictionary synthesis, and schema artifact production
• Produce data dictionaries, ERD diagrams, ETL field mapping specifications, and integration contracts
• Collaborate with SMEs to validate models and integration flows against undocumented business logic

Required Qualifications:

• 8+ years experience in data architecture and system integration within OLTP / transactional domains (insurance, banking, billing, or similar)
• Hands-on experience with IBM IMS, DB2 for z/OS, and COBOL copybooks able to read a segment hierarchy or copybook independently
• 5+ years designing physical data models for modern relational or document-oriented databases
• Strong grasp of logical modeling: ERD notation, composition vs. reference, cardinality, key design
• Proven experience designing integration architectures: REST APIs, event streaming (Kafka, MQ), CDC pipelines, and message-based data distribution
• Experience with ELT processing, including designing and implementing ELT workflows, data transformation, data cleansing, and data validation
• Experience with real-time data processing, including designing and implementing real-time data processing pipeline with event-driven architectures
• Comfort working in an AI-augmented workflow directing LLM-based agents, reviewing AI-generated artifacts, and applying domain expertise to close gaps AI cannot resolve
• Scripting proficiency (Python or equivalent) for schema validation and artifact generation
• Ability to abstract legacy physical data structures into business-oriented target models separating IMS/DB2 storage implementation details from true business keys and domain entities

Preferred

• Insurance domain knowledge (policy lifecycle, coverages, LOBs, premium rating)
• Domain-Driven Design (DDD) - aggregates, bounded contexts, event-driven design
• Experience synthesizing a unified model from multiple heterogeneous sources (IMS + DB2 + application logic)
• Prior experience working with AI coding assistants (Claude, GitHub Copilot, or similar) in a software engineering or data architecture context