Senior Database Architect

WEX•Portland, ME

About The Position

We are seeking a Senior Database Architect who combines deep expertise in legacy database systems with forward-looking vision for AI-native data architecture. You'll lead the decomposition of complex stored procedures while simultaneously designing the vector databases, embedding strategies, and semantic models that power our AI agents and workflows. This is an AI-first role in two senses: you'll leverage AI to accelerate your own work (stored procedure analysis, migration generation, schema documentation), and you'll design the data infrastructure that AI systems depend on. If you're excited about both solving hard legacy database problems and architecting the data layer for AI-native applications, this role is for you.

Requirements

8–12 years in database engineering and architecture, with significant experience in enterprise-scale SQL Server environments
Deep SQL Server expertise: T-SQL optimization, stored procedure design and refactoring, query plan analysis, indexing strategies, and performance tuning
Hands-on modernization experience: track record of decomposing complex stored procedures and migrating business logic to application services
Multi-platform data architecture: experience designing solutions across relational (SQL Server, PostgreSQL), NoSQL (MongoDB, Cosmos DB), and analytical (Snowflake, data lakehouse) platforms
Event-driven data patterns: CDC, Kafka, outbox pattern, event sourcing, CQRS—practical experience implementing these in production
Data modeling expertise: canonical models, dimensional modeling, schema evolution, and designing for extensibility
Vector database experience: hands-on with at least one vector DB (Pinecone, Weaviate, Milvus, pgvector, Azure AI Search, or similar)
RAG architecture understanding: embedding models, chunking strategies, retrieval optimization, hybrid search, and reranking patterns
Semantic modeling: experience designing data structures optimized for AI retrieval—knowledge representation, ontologies, or domain-specific schemas for AI consumption
Understanding of embedding pipelines: text preprocessing, embedding generation, vector indexing, and incremental updates
Familiarity with LLM context requirements: what data AI agents need, token constraints, context window optimization
2+ years actively using AI coding assistants for database work; deep understanding of how to prompt effectively for SQL and data engineering tasks
Experience building tools, scripts, or automation that leverage AI/LLM capabilities
Familiarity with structured artifact creation for AI consumption: documented schemas, annotated procedures, context files
Vision for AI-assisted database engineering and ability to build tooling that enables it
Strong programming skills in at least one backend language (C#, Java, Python) for building migration tooling, embedding pipelines, and services
Cloud data services experience: Azure SQL, Cosmos DB, Azure AI Search, Azure Synapse, Snowflake, or AWS equivalents
Infrastructure-as-code for data platforms: Terraform, ARM/Bicep, or CloudFormation
Understanding of domain-driven design and how data architecture supports bounded contexts
Familiarity with data governance, lineage, and compliance requirements (HIPAA, PCI-DSS)

Nice To Haves

Background in healthcare, benefits, payments, or similarly regulated industries
Experience building RAG systems or AI-powered search/retrieval applications
Knowledge graph experience: Neo4j, Amazon Neptune, or similar graph databases
Contributions to database tooling, AI/ML data infrastructure, or open-source projects
Experience mentoring engineers or leading database/data architecture communities of practice

Responsibilities

Analyze and decompose large SQL Server stored procedures (1,000+ lines) with embedded business logic, creating migration strategies that extract logic into domain services
Design patterns for separating business rules from data access, enabling stored procedures to become thin data-access layers while business logic moves to application services
Lead refactoring efforts that align database structures with domain-driven design: bounded contexts, aggregates, and domain events
Implement event-driven patterns that decouple systems from direct database dependencies: change data capture, outbox patterns, event sourcing where appropriate
Optimize query performance, indexing strategies, and execution plans as part of modernization efforts
Create migration playbooks and tooling that engineering teams can apply to their own stored procedure modernization
Design semantic data models that capture domain knowledge in structures optimized for AI retrieval and reasoning
Architect vector database solutions for RAG implementations: embedding strategies, chunking approaches, similarity search optimization, and hybrid retrieval patterns
Design and implement embedding pipelines that transform domain content into vector representations suitable for AI agent consumption
Establish knowledge graph patterns where appropriate: entity relationships, ontologies, and graph-based retrieval for complex domain reasoning
Define data architectures for AI agent context: what data agents need, how it's structured, how freshness and consistency are maintained
Design evaluation frameworks for RAG quality: retrieval accuracy, relevance scoring, and feedback loops for continuous improvement
Design canonical data models and schemas that are flexible, extensible, and aligned with business domain concepts
Architect data solutions across multiple platforms: SQL Server, PostgreSQL, MongoDB/Cosmos DB, Snowflake, and vector databases (Pinecone, Weaviate, pgvector, Azure AI Search)
Design event-driven data flows: Kafka-based event streaming, materialized views, CQRS patterns, and real-time data synchronization
Establish data platform infrastructure patterns: data pipelines, ETL/ELT orchestration, data quality frameworks, and observability
Define data residency, partitioning, and multi-region strategies for performance and compliance
Create reference architectures for common data patterns that domain teams can adopt
Leverage AI coding assistants (GitHub Copilot, Cursor, Claude Code) to accelerate stored procedure analysis, refactoring, and migration
Build AI-powered tools for database engineering: automated stored procedure analysis, schema documentation generators, migration assistants, and query optimization recommenders
Create AI-consumable artifacts: structured documentation, annotated schemas, and context files that enable AI agents to understand and work with database systems
Author database architecture skills that encode patterns, constraints, and best practices for AI-assisted development
Develop prompts, workflows, and tooling that help engineering teams apply AI effectively to database modernization tasks
Partner with AI/ML teams to ensure data architecture supports agent and workflow requirements
Collaborate with domain teams to understand their data requirements and design solutions aligned with domain ownership
Work with application architects to ensure data architecture supports service-oriented and event-driven designs
Contribute to Enterprise Architecture Council (EAC) standards for data architecture, modeling conventions, and technology selection
Mentor engineers on database design, optimization, semantic modeling, and AI data infrastructure