DIMP Model Engineer

Risk Model Engineer - Gas Distribution Integrity

Location: Remote (US or Canada)

Department: Insights (AI/ML)

Reports to: Data Science Team Lead

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 never previously lived in one place 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 that data.

We are hiring a Risk Model Engineer to own the gas distribution risk models within that analytical layer.

Distribution is a new model domain for the platform. The threat framework and model set are being defined now in collaboration with subject-matter experts and design-partner operators. In this role, you will help define the models, build them, and take them through validation-producing outputs that utilities can use to prioritize replacement programs and support regulatory proceedings.

The modeling work focuses on large populations of buried assets with relatively few observed failures, incomplete records, and limited direct inspection data. The work spans physics-based probabilistic models and machine learning, with an emphasis on producing risk rankings that remain explainable, calibrated, and defensible.

You will own the model content: what the model computes, why it computes it, and the evidence supporting its validity.

Productionization-including model serving, data pipelines, deployment, and monitoring infrastructure-is owned by dedicated ML Ops and data engineering roles with whom you will work closely.

Key Responsibilities

1. Threat Framework & Integration

• Translate the gas distribution threat framework, in collaboration with subject-matter experts, into systematic threat identification methods and supporting data structures.
• Integrate model inputs from diverse datasets across Irth's product offerings.
• Identify gaps, inconsistencies, and limitations in source data and determine how they should be reflected in model inputs and outputs.

2. Model Development

• Extend threat coverage across domains, including:
• Legacy material corrosion
• Plastic embrittlement
• Excavation damage
• Cross-bores
• Other gas distribution