AI Application Engineer

The Position

We're building enterprise AI platforms that automate document-intensive review workflows across the business. The AI does the heavy lifting on document understanding, rule validation, and exception detection. The Product Engineer builds what users actually see and use — the reviewer workbench, the executive reporting layer, the data products, and the audit-ready outputs that turn AI insights into business decisions.

About the Role

This is a hybrid product engineering + analytics engineering role. You'll partner with the AI Solution Architect (who owns the agents) and the MLOps Engineer (who owns the data platform) to ship the user-facing layer. You report to the Head of AI & Data Strategy.

Responsibilities

• Reviewer Workbench & application engineering
  • Reviewer-facing application — triage incoming documents, accept or override AI recommendations, annotate decisions
  • UI/UX patterns that match how reviewers actually work today (comment logs, page-level evidence links, override flows)
  • Feedback capture — every reviewer interaction trains the model and refines the rule packs
  • Application performance at scale — handling high document volumes with thousands of pages of supporting context
  • Surfacing of agent outputs, confidence scores, and citations to reviewers in a way they trust
• Reporting, dashboards & executive visibility
  • Power BI dashboards on Databricks SQL Warehouse — review throughput, accuracy, time saved, operational metrics
  • Executive reporting layers for SLT and business leadership
  • Portfolio analytics — cross-domain benchmarks, trend analysis, anomaly surfacing
  • Quarterly reporting infrastructure for SteerCo and AI Program Council reviews
• Data products & analytics engineering
  • Clean, documented Gold-layer data products on Unity Catalog — reusable across business stakeholders, IT Delivery, AI team
  • Output generators — auto-produce Excel + Word + linked evidence outputs per reviewed item, audit-ready by default
  • APIs and integration patterns surfacing platform outputs to enterprise systems and business workflows
  • Analytics engineering standards — dbt or equivalent modeling, version control, lineage, documentation
• User research & adoption
  • Direct work with business users — what's slowing them down, what's working, what they need next
  • Fast-cadence iteration based on real user feedback, not quarterly release cycles
  • User documentation, training materials, onboarding flows for end users and report consumers
  • Cross-functional partnership with the AI Solution Architect, MLOps Engineer, business stakeholders, IT Delivery, SLT

Must-have Qualifications

• 5+ years building data-intensive applications, internal tools, or data products in production
• Full-stack capability — modern web frameworks (React, Vue, or equivalent), Python or TypeScript, SQL
• Production experience with at least one BI/reporting tool (Power BI preferred)
• Strong SQL on Databricks SQL Warehouse, dbt, or equivalent lakehouse modeling
• Proven UX intuition — you have shipped internal tools real users adopt and improved them based on observed behavior
• Experience designing reporting consumed by executive stakeholders — balancing detail, accuracy, clarity
• Microsoft 365 and Azure ecosystem familiarity (Power BI, SharePoint Online, Azure App Services, Entra ID)
• Bachelor's degree in Computer Science, Engineering, or related field

Nice to Have

• Power BI Data Analyst Associate or Databricks Data Analyst Associate certifications
• Document-heavy UX experience — PDF viewers, annotation tools, evidence linking, comment logs
• Experience surfacing LLM outputs to end users — confidence scoring, override flows, citation interfaces
• React + TypeScript + modern frontend tooling and component libraries
• Embedded analytics (Power BI embedded, Looker embedded, or equivalent)
• Construction, real estate, data center, or AEC industry context
• Experience operating in a small AI team where you wear multiple hats

How You Work

• You build for the user in front of you. You would rather watch a user use what you built than read about adoption metrics.
• You're comfortable with ambiguity. You figure out what the user needs by talking to them, not by waiting for a spec.
• You ship small and often. Big quarterly releases are not your model.
• You think about data products as products. A Gold-layer table should be documented, versioned, and reliable.

Pay

$128,260.00 - $146,017.59

Benefits

• Healthcare, Dental Care, Vision Insurance, Life Insurance, Paid Time Off, and Paid Leave Programs
• 401K program, flexible spending accounts, cell phone subsidy
• Peer-to-peer recognition and rewards programs
• Events, game nights, happy hours, and barbecues

Schedule & Location

Denver, CO. Travel under 10%. Must be eligible to work in the United States. Must pass comprehensive background and drug screening.