Technical Product Manager

About this role

About Irth Irth is building Sentinel , a unified data and analytics platform that evolves our Asset Integrity for Pipelines product from a point risk-scoring feature into a comprehensive threat-management platform. Sentinel connects asset integrity, damage prevention, stakeholder management, and land management. The platform will support industry-standard, Irth-built, and customer-built ("bring your own model") risk models across both transmission and distribution integrity management programs. We are hiring a Technical Product Manager dedicated full-time to Sentinel . This is a hands-on product role at the intersection of data engineering, applied analytics, and regulated workflow software. You will own product requirements and delivery across a multi-phase roadmap that includes: Building a unified data platform with AI-driven ETL Decoupling risk models into a standalone analytical layer Launching a model governance and BYOM (Bring Your Own Model) framework Expanding Sentinel from transmission into distribution use cases The Product Challenge The defining challenge of Sentinel is not the models — it is the data . The industry risk models we support require approximately 65 input attributes. Our products natively cover most, but not all, of these attributes, while customer data arrives in many forms, including spreadsheets, PDFs, inspection reports, and field notes. Your core product challenge will be closing that data gap through automated ingestion, normalization, quality repair, and reconciliation , rather than relying on services-heavy customer onboarding. We are open on level. Title and scope will be calibrated to the experience you bring. Strong candidates at both the mid- and senior-level individual contributor range are encouraged to apply. What You'll Do Own Product Definition Translate a phased product strategy and high-level requirements into a groomed, traceable backlog of epics, features, user stories, and acceptance criteria that engineering can execute against without ambiguity. Drive discovery with design partners and prospective customers, translating operational and regulatory realities into clear product requirements. Maintain the roadmap against defined phase gates and hold the team accountable to delivery criteria, including: Data platform operational Data lineage traceable Risk models decoupled BYOM framework live Distribution capabilities generally available Lead the Data & Analytics Workstream Partner with data architects, data engineers, and data scientists to define the unified data platform, ingestion and normalization layer, and data-quality capabilities. Define requirements for automated data-quality repair, drift monitoring, and reconciliation. Specify connectors to customer systems of record and external data providers, including weather, soil, satellite imagery, and locate-ticket sources. Define entity-resolution requirements and shared spatial data capabilities so records can be reconciled to common assets across product lines. Own requirements for model configuration, versioning, governance, and audit trails across industry, Irth, and customer-built models. Make Agentic AI a Delivery Advantage Shape agentic AI workflows for document parsing and schema mapping. Scope proofs of concept and define human-in-the-loop review thresholds and quality standards for production deployment. Use AI tooling as part of your own product development workflow. Irth uses Claude Code for agentic software development, and we expect Product to work alongside Engineering in that environment rather than simply handing requirements over the wall. Deliver in a Regulated Environment Translate federal pipeline safety regulations into executable product logic, auditable outputs, and defensible reporting. Define requirements for traceability, compliance documentation, and outputs suitable for utility rate-case filings. Define validation and sensitivity-testing requirements for every model release before commercial launch. Partner with Engineering and Security to establish the non-functional product bar, including: Multi-tenant isolation Access control Calculation reproducibility Auditability Security-attestation readiness Run the Delivery Cadence Drive agile execution across a cross-functional team, including sprint planning, dependency management, release discipline, and delivery metrics. Coordinate across data, application, GIS, and QA workstreams and proactively resolve cross-team dependencies. Partner with Product Marketing, Sales Engineering, and Account teams on design-partner enablement, pilot feedback loops, and launch readiness. What Success Looks Like You will help Sentinel move from a point risk-scoring capability to a scalable threat-management platform that can: Reliably ingest and normalize complex customer data Reduce manual, services-heavy onboarding Provide traceable and trusted data lineage Support multiple risk-model frameworks Enable customers to bring and govern their own models Deliver auditable, reproducible analytical results Expand from transmission into distribution integrity management Use AI to materially improve data onboarding and product delivery Requirements What You'll Bring Required Qualifications 4+ years of experience in technical product management, product ownership, or a closely related role delivering enterprise software, ideally where data quality, analytics, or modeling was central to the product's value proposition. Demonstrated ability to translate data science and statistical modeling concepts into governed, structured, and actionable end-user workflows, and to communicate model behavior effectively to non-technical stakeholders. Hands-on fluency with data, including schema design and alignment, pipeline and ingestion design, handling missing or inconsistent data, and governance practices that make outputs defensible and audit-ready. Practical experience applying AI to product and delivery workflows, including prompt engineering, AI-assiste