AI/ML Data Architecture & Governance Leader

Matthew Stevens

*************@*******.*** +1-423-***-**** https://www.linkedin.com/in/matthew-stevens-engineer/ Professional Summary

AI/ML Data Architect with 13 years of experience designing enterprise data platforms, lakehouse architectures, cloud-native analytics ecosystems, AI-ready data foundations, GenAI platforms, semantic layers, MDM strategies, and governed data products across restaurant technology, eCommerce, insurance, financial workflows, hospitality, and enterprise SaaS environments. Deep experience defining architecture blueprints for modern data platforms using Snowflake, Databricks, Unity Catalog, BigQuery, Azure Data Lake, Azure Databricks, Delta Lake, Apache Iceberg, Redshift, AWS, Azure, GCP, Terraform, Airflow, Dagster, Prefect, dbt, Spark, Kafka, Flink, and API-driven integration patterns. Strong background architecting data solutions that support analytics, business intelligence, AI/ML, RAG systems, intelligent assistants, agentic workflows, vector search, semantic search, feature engineering, model training pipelines, real-time inference, and responsible AI governance. Proven ability to design enterprise data domains, canonical data models, master data management patterns, metadata frameworks, lineage controls, data contracts, access-control standards, and governance-by- design workflows that improve trust, reuse, discoverability, and regulatory readiness. Experienced in guiding cross-functional teams through modernization initiatives, legacy migration, cloud platform adoption, multi- cloud strategy, architecture review boards, technical design documentation, platform reliability standards, and cost-performance optimization. Known as a hands-on architecture leader who can translate business strategy into practical solution blueprints, mentor engineers, evaluate emerging technologies, build reference implementations, and deliver scalable platforms that make data reliable, secure, AI-ready, and usable across the enterprise.

WORK EXPERIENCE

Lead Engineer/AI,ML Data Architect Culver Franchising System Remote Mar 2020 - Present

● Led the architecture and evolution of a cloud-native restaurant, eCommerce, franchise operations, and analytics data platform supporting digital ordering, menu pricing, inventory operations, payment-adjacent workflows, store operations, customer engagement, AI-assisted support, and enterprise reporting across multi-region environments.

● Defined the target-state data architecture blueprint using lakehouse, warehouse, medallion, semantic-layer, and domain-oriented data product patterns to improve scalability, governance, self-service analytics, and AI/ML readiness across operational and analytical workloads.

● Designed and implemented AI-ready data pipelines using Python, SQL, Spark, PySpark, Databricks, Snowflake, BigQuery, Redshift, dbt, Airflow, Dagster, AWS Glue, Azure Data Factory-style patterns, and cloud-native orchestration to move data from raw source systems into curated business-ready layers.

● Architected Bronze, Silver, and Gold data models in Databricks and cloud warehouse environments, creating reliable schemas for orders, stores, menu items, inventory, payment status, customer behavior, franchise reporting, operational events, and AI-powered support use cases.

● Designed GenAI and RAG architectures using AWS Bedrock, OpenAI/Anthropic-style LLMs, embeddings, vector search, semantic search, LangChain/LlamaIndex patterns, prompt governance, retrieval evaluation, and controlled response workflows for internal support and operational knowledge assistants.

● Built AI evaluation and responsible AI control loops that tracked prompt versions, retrieval sources, response confidence, hallucination risk, user feedback, latency, token cost, and answer quality to make AI- assisted workflows safer, more explainable, and more trustworthy for business users.

● Established metadata, lineage, and governance practices using catalog-driven design patterns similar to Unity Catalog, DataHub, Dataplex, OpenMetadata, Atlan, and enterprise data catalog frameworks to improve discoverability, ownership, auditability, and reuse of key datasets.

● Designed automated data quality frameworks using validation rules, schema enforcement, reconciliation checks, anomaly detection, freshness monitoring, completeness checks, regression testing, Pydantic-style validation, and CI/CD-driven quality gates to reduce downstream reporting issues.

● Implemented governance-by-design patterns around data contracts, domain ownership, RBAC, row-level and column-level access controls, catalog separation, environment isolation, secure storage boundaries, audit logging, retention rules, and privacy-conscious handling of customer and transaction-related data.

● Modernized legacy reporting and operational data flows into reusable ETL/ELT frameworks with batch, streaming, CDC-style, API-based, and event-driven ingestion patterns using Kafka, Kinesis, SQS, SNS, EventBridge, Pub/Sub-style messaging, and replayable event streams.

● Architected integration middleware and Backend-for-Frontend patterns that allowed customer-facing applications, internal dashboards, and operational services to consume governed APIs and cached data access layers instead of directly querying raw warehouse or lakehouse resources.

● Optimized cloud data platform performance and cost by improving partitioning, clustering, file layout, query execution plans, warehouse sizing, compute scaling, caching, materialized views, storage lifecycle policies, and tenant/team-level cost attribution models.

● Evaluated and introduced open table format concepts including Delta Lake, Apache Iceberg, Hudi, Nessie- style catalog/versioning ideas, schema evolution, time travel, zero-copy cloning, and interoperability patterns across Snowflake, Databricks, BigQuery, Spark, and Trino-style query engines.

● Partnered with product, operations, finance, analytics, data science, security, DevOps, and executive stakeholders to translate ambiguous business goals into architecture roadmaps, solution blueprints, technical design documents, data models, and measurable delivery plans.

● Designed AI/ML enablement patterns for demand forecasting, inventory optimization, store performance analytics, customer engagement, support automation, anomaly detection, feature engineering, model-input monitoring, and curated training datasets.

● Used Terraform, CloudFormation-style infrastructure-as-code, GitHub Actions, Cloud Build-style pipelines, Docker, Kubernetes, CI/CD, automated testing, and environment promotion standards to make data platform infrastructure repeatable, secure, and version-controlled.

● Established observability standards for data systems using structured logs, lineage-aware alerts, pipeline health metrics, SLA/SLO tracking, warehouse cost monitoring, model-data freshness checks, runbooks, and root-cause analysis practices.

● Built AI-assisted engineering workflows using Claude Code, Cursor, GitHub Copilot, prompt templates, custom commands, automated documentation, pipeline scaffolding, test generation, and architecture-review support while enforcing code quality and human review standards.

● Mentored data engineers, backend engineers, and analytics partners through architecture reviews, design patterns, data modeling standards, AI/ML solution reviews, governance practices, pipeline design, production readiness, and platform modernization decisions.

● Served as a hands-on technical anchor for strategic data and AI decisions, balancing business urgency, engineering rigor, platform cost, data governance, user adoption, and long-term maintainability. Senior Software Engineer/AI,ML Data Architect Allstate Insurance Co.Northbrook, IL Apr 2018 - Dec 2019

● Designed secure insurance and financial data architectures supporting policy management, claims intake, billing operations, underwriting workflows, document processing, customer-service tools, partner integrations, reconciliation, compliance reporting, and enterprise analytics.

● Built and modernized cloud-native data pipelines using Python, SQL, Spark, PySpark, AWS Glue, Airflow, Redshift, Snowflake, BigQuery, PostgreSQL, DynamoDB, MongoDB, S3, Kafka, RabbitMQ, SQS, SNS, and API-based ingestion patterns for structured and unstructured insurance datasets.

● Architected source-to-target mappings, transformation logic, metadata structures, serving layers, and curated reporting models for claims, policies, billing records, customer profiles, audit events, document metadata, workflow status, and operational performance.

● Designed dimensional models, star schemas, normalized models, canonical business entities, and curated data marts to support claims analytics, finance reporting, compliance review, customer operations, service dashboards, and data science experimentation.

● Implemented governance frameworks for regulated data using metadata tagging, lineage documentation, sensitive-field classification, audit trails, data retention policies, access controls, least-privilege permissions, and privacy-aware data movement.

● Supported insurance-domain AI/ML use cases by preparing feature-ready datasets, model-training extracts, claims-routing inputs, risk-scoring data, document classification datasets, fraud-pattern signals, and evaluation datasets for analytics and machine learning teams.

● Designed RAG-ready retrieval layers for policy documents, claims records, billing knowledge, customer- service procedures, and operational documentation using search indexes, metadata extraction, embeddings, semantic retrieval patterns, and controlled API outputs.

● Implemented automated data quality and reconciliation workflows to compare source records, transformed datasets, downstream reports, partner files, billing outputs, claims-processing results, and audit datasets, improving confidence in regulated reporting.

● Built batch and event-driven workflows for claims events, billing status updates, document intake, customer notifications, partner data exchange, downstream synchronization, and operational reporting using reliable retry, backfill, dead-letter, and failure-recovery patterns.

● Improved query and platform performance through indexing, partitioning, query-plan review, caching, warehouse optimization, read-path redesign, data-access refactoring, and transaction-boundary improvements, reducing key data-service response times by 40%.

● Designed secure API and integration architectures for internal portals, partner systems, customer-facing workflows, third-party services, and downstream analytics consumers using REST, GraphQL, OAuth2, JWT, structured validation, and audit-friendly contracts.

● Applied Secure SDLC, HIPAA-conscious data practices, SOC 2-style auditability, HITRUST-aware documentation concepts, PII protection, encryption-aware configuration, OWASP controls, and compliance- aligned release validation across data and application workflows.

● Collaborated with product owners, compliance partners, finance users, claims operations, QA, architecture, support, and data science teams to translate complex business needs into scalable data architecture designs and production-ready implementation plans.

● Authored data-flow documentation, architecture decision records, technical design notes, integration diagrams, governance recommendations, validation plans, and operational runbooks to support long-term maintainability and cross-team alignment.

● Helped modernize legacy insurance reporting and data-processing services into modular, cloud-native, observable data pipelines with clearer ownership, reusable patterns, stronger quality controls, and better production support practices.

● Supported CI/CD and DevOps practices using GitHub, Jenkins, CircleCI-style pipelines, Dockerized test environments, automated regression checks, deployment documentation, environment promotion, and rollback planning for data and application releases.

● Participated in architecture reviews, production readiness reviews, code reviews, incident analysis, root- cause investigations, and platform improvement discussions to raise reliability, data integrity, and system scalability.

● Mentored junior engineers on data modeling, SQL optimization, pipeline orchestration, secure data handling, API integration, cloud architecture, documentation, and stakeholder communication.

● Served as a bridge between business stakeholders and technical teams, helping clarify tradeoffs around compliance, performance, cost, timeline, data quality, and long-term architecture strategy. Data Architect / Full Stack Developer Comfort Inn North Bethesda, MD Jun 2013 - Oct 2017

● Developed hospitality, property-management, reservation, reporting, and operational data applications supporting guest profiles, check-in workflows, room availability, service requests, payment-status tracking, vendor integrations, mapping, and back-office operations across distributed hotel environments.

● Designed early-stage data models and operational data structures for guests, reservations, rooms, service requests, payment references, property assets, audit records, location metadata, support notes, vendor activity, and reporting datasets.

● Built ETL and reporting workflows using Python, SQL, MySQL, PostgreSQL, MongoDB, Redis, Azure SQL Database, Azure Storage, and scheduled processing patterns to support occupancy analytics, reservation reporting, service trends, and operational KPIs.

● Created API-driven integration patterns for reservation systems, property-management screens, vendor services, map platforms, customer-facing pages, internal dashboards, guest communications, and reporting modules.

● Helped migrate legacy monolithic functionality into more modular service-oriented systems, reusable API layers, and structured data access patterns, improving maintainability, scalability, deployment flexibility, and onboarding speed.

● Designed batch processing and asynchronous workflows for booking notifications, guest-service updates, operational alerts, reporting refreshes, vendor synchronization, map-related updates, and support escalations.

● Integrated external data providers and third-party services using REST APIs, JSON, XML, secure callbacks, WebSocket-style updates, standardized payload contracts, and retry-safe processing patterns.

● Built analytical queries, scheduled extracts, internal dashboards, and operational reports for reservation volume, occupancy trends, service requests, guest support activity, payment-status exceptions, vendor performance, and property operations.

● Improved application and reporting performance by 50% through SQL query optimization, index tuning, schema cleanup, caching, connection management, removal of duplicate logic, and refactoring inefficient data-access flows.

● Supported cloud-hosted application and data delivery using AWS-hosted environments, Azure App Service, Azure SQL Database, Azure Storage, Git workflows, deployment scripts, environment configuration, and monitoring practices.

● Implemented data validation rules, duplicate checks, audit records, error logs, operational controls, and reconciliation routines to improve accuracy across reservation records, room assignments, guest profiles, and reporting outputs.

● Developed location-aware data workflows using Google Maps API and Esri Maps API to support property visibility, service-area reporting, geographic analysis, and operational planning.

● Supported payment-adjacent hospitality workflows with secure session handling, RBAC concepts, audit logging, sensitive-data minimization, privacy-aware guest-data handling, and web security practices.

● Created documentation for data models, API contracts, deployment steps, operational workflows, troubleshooting guides, and support handoffs to improve communication between development, QA, support, and hotel operations teams.

● Participated in Agile delivery activities including requirements review, estimation, development, testing, bug fixing, release support, stakeholder demos, production troubleshooting, and post-release validation.

● Worked closely with product owners, hotel operations teams, QA engineers, support staff, vendors, and senior developers to translate operational problems into practical data-backed solutions.

● Built a strong foundation in data modeling, API architecture, reporting automation, SQL optimization, integration design, stakeholder communication, and production support that later evolved into enterprise- scale AI and data architecture leadership.

EDUCATION

Columbia College Chicago

Bachelor of Computer Science Sep 2008 - Jun 2012

SKILLS

● Data & AI Architecture: Enterprise data architecture, AI/ML data architecture, GenAI architecture, RAG architecture, agentic workflows, intelligent automation, responsible AI, prompt governance, model evaluation, AI governance, AI-ready data platforms, feature engineering, feature stores, model-training pipelines, real-time inference data flows, AI/ML lifecycle enablement, LLMOps, MLOps, analytics architecture, data product architecture, data mesh, data fabric, domain-driven data design, semantic layers, canonical data models, enterprise data models, MDM, master data management, Profisee concepts, source- of-truth strategy

● Cloud Data Platforms: AWS, Azure, GCP, Snowflake, Databricks, Unity Catalog, BigQuery, Redshift, Azure Data Lake Storage Gen2, Azure Blob Storage, Azure Databricks, Azure Data Factory, Azure Synapse Analytics, Microsoft Fabric, Azure Functions, Azure Event Hubs, Azure Stream Analytics, Azure Kubernetes Service, AWS S3, AWS Glue, AWS Lambda, AWS RDS, Aurora PostgreSQL, DynamoDB, EMR, Athena, Kinesis, MWAA, DMS, CloudWatch, IAM, VPC, KMS, Secrets Manager, Google Cloud Dataflow, Dataproc, Pub/Sub, Dataplex, Data Catalog, Cloud Composer, Vertex AI

● Lakehouse, Warehouse & Storage Architecture: Lakehouse architecture, data warehouse architecture, data lake architecture, Delta Lake, Apache Iceberg, Apache Hudi, Nessie concepts, open table formats, zero- copy cloning, schema evolution, ACID table formats, compaction, data lifecycle management, Snowflake Cortex concepts, Databricks Lakehouse, SQL Server, Oracle, PostgreSQL, MySQL, MariaDB, MongoDB, DynamoDB, Cosmos DB, NoSQL, graph databases, vector databases, object storage, file storage, block storage, archive storage, hybrid storage architecture

● Data Engineering & Orchestration: ETL, ELT, CDC, batch processing, streaming processing, near-real- time processing, event-driven architecture, API-based ingestion, data integration middleware, BFF patterns, orchestration, workflow dependency management, backfills, retries, idempotency, Airflow, Dagster, Prefect, dbt, Azure Data Factory, Synapse Pipelines, Apache Beam, Dataflow, Apache Spark, PySpark, Flink, Kafka, Confluent, Kinesis, RabbitMQ, SQS, SNS, Pub/Sub, Event Hubs, Apache NiFi, Hadoop, HDFS, Hive, HBase, Trino, Starburst, Presto, Athena

● Data Modeling & Governance: Conceptual data modeling, logical data modeling, physical data modeling, dimensional modeling, star schema, snowflake schema, 3NF, Data Vault, normalized modeling, schema design, source-to-target mapping, metadata management, data lineage, data cataloging, data discovery, data contracts, schema registry, data quality standards, data ownership, stewardship, data lifecycle policies, RBAC, ABAC, row-level security, column-level security, data classification, auditability, traceability, data provenance, model provenance, policy enforcement, governance-by-design

● AI/LLM & Intelligent Automation: OpenAI API, Anthropic Claude, Claude Code, Cursor, GitHub Copilot, AWS Bedrock, Azure ML, Azure AI Foundry concepts, Vertex AI, LangChain, LlamaIndex, LangGraph concepts, embeddings, vector search, semantic search, RAG pipelines, prompt engineering, prompt versioning, AI assistants, AI agents, custom slash commands, AI-assisted development, AI-powered documentation, AI-assisted data quality, Dataiku concepts, UiPath/RPA architecture concepts, intelligent process automation

● Observability, Quality & FinOps: Data observability, Monte Carlo concepts, Dynatrace concepts, Atlan concepts, Great Expectations-style validation, Pydantic v2 validation, schema checks, freshness checks, completeness checks, anomaly detection, regression detection, reconciliation, SLA/SLO tracking, cost allocation, FinOps, tenant-level cost attribution, token usage tracking, warehouse spend optimization, compute scaling, pipeline monitoring, structured logging, runbooks, root-cause analysis, incident response, production readiness reviews

● Infrastructure, DevOps & Architecture Practices: Terraform, CloudFormation, AWS CDK concepts, infrastructure as code, GitLab CI, GitHub Actions, Jenkins, CircleCI, Azure DevOps, Docker, Kubernetes, Helm, AKS, EKS, GKE concepts, CI/CD for data workflows, automated database deployment, reproducible environments, DevSecOps, architecture review boards, design reviews, technical design documents, architecture decision records, reference architectures, solution blueprints, C4 diagrams, TOGAF concepts, AWS Well-Architected Framework

● Security, Compliance & Regulated Domains: HIPAA, HITRUST, SOC 2, PCI-DSS, FINRA, SEC, NIST, FHIR, HL7, EHR/EMR data exchange, healthcare data interoperability, payer/provider data, insurance policy/claims/billing data, financial services data, privacy-by-design, encryption at rest, encryption in transit, secrets management, IAM governance, network isolation, Zero Trust access patterns, data retention, compliance-aligned architecture, secure API integration

● Business & Leadership: Principal-level architecture, staff-level technical leadership, enterprise roadmap planning, data strategy, AI strategy, modernization strategy, architecture governance, stakeholder influence, executive communication, vendor evaluation, proof of concept leadership, reference implementation, team mentoring, architecture documentation, backlog shaping, roadmap prioritization, Agile, Scrum, Kanban, cross-functional collaboration, engineering standards, platform modernization, legacy migration, cost- performance tradeoff analysis

Contact this candidate