Data Architect (Mid-Level)

Data Architect – Insights (AI/ML) Location: Remote (US) Department: Insights (AI/ML) Reports to: Engineering Manager About the Role Irth is building a new AI-driven threat and risk management platform for pipeline asset integrity. The platform brings together three capabilities that have historically been separate at Irth: - A governed, cross-product data platform built on Databricks and Azure - An AI-powered ingestion layer that normalizes, repairs, and enriches customer data without services-heavy onboarding - A reusable analytical layer that runs industry-standard, Irth-developed, and customer-built risk models against the data We are seeking a mid-level Data Architect to design the schema and data model that powers this platform. You will create a unified structure spanning asset integrity, damage prevention, and land and stakeholder management, capable of supporting probabilistic risk models that require dozens of precisely defined input attributes per pipeline segment. This is a hands-on design role. You will establish standards for data lineage, governance, and quality while working closely with the data engineering team to implement them. This role is ideal for someone with substantial experience in lakehouse and medallion architecture who is ready to own the data model for a platform rather than a single project. Key Responsibilities 1. Data Modeling & Architecture - Design the schema and unified data model for the platform, spanning asset, inspection, incident, geospatial, and consequence data across multiple products. - medallion architecture patterns (Bronze, Silver, Gold) and define the semantic layer consumed by downstream models, reporting, and applications. - Model the data required by probabilistic risk models, ensuring every required input attribute has a defined source, transformation path, and data-quality expectation. - Design cross-product entity resolution so records from separate products reconcile to d assets and entities. - Define ingestion patterns for batch, streaming, and change-data-capture (CDC) sources, including schema evolution and slowly changing dimensions. - Produce architecture documentation, diagrams, data dictionaries, and decision records that engineers can implement without ambiguity. 2.

Geospatial & Domain Data Design - Design the d GIS layer as a first-class component of the data model, including pipeline centerlines, consequence-area polygons, right-of-way corridors, and parcel data. - Model linear referencing and dynamic segmentation so risk results align with centerline geometry and can be aggregated across differing segmentations. - Define data structures for inspection data, repair history, and external enrichment sources such as weather history, soil characteristics, and satellite-derived data. 3. Governance, Lineage & Data Quality - Define cataloging, classification, and metadata standards in Unity Catalog and establish expectations for lineage coverage across the data estate. - Establish data-quality rules, validation gates, and profiling standards, including processes for quarantining and remediating failures. - Define access-control patterns, including role-based and attribute-based access, tenant isolation, and appropriate handling of sensitive data. - Design for end-to-end traceability so that any risk output can be traced through every transformation back to its source record. 4. Implementation Partnership - Work directly with data engineers to translate architectural designs into production pipelines and review implementations for conformance. - Contribute hands-on to schema deployment, transformation logic, and optimization when it accelerates delivery. - Partner with application engineers and data scientists to determine how the data model is consumed through serving layers, feature stores, and APIs. - Support performance and cost optimization, including partitioning, clustering, file layout, and compute sizing. 5.