AI Engineer - Data Intelligence

The Opportunity

AI-powered platforms, like Clarium’s, deliver the highest impact when they are supported by high-quality data. As we scale to more health systems and deepen our offering of intelligent, data-driven workflows, the master data enrichment pipeline (the system that classifies and contextualizes every product flowing through a hospital's supply chain) has become a critical growth lever. We're investing in the team and infrastructure to make that layer faster, smarter, and more reliable.

Data Products Team

The Data Products team is a small, unusually senior group responsible for the data assets, data science, and analytics that drive measurable value for our clients. You'll join the team, building and owning components of our enrichment pipeline: classification workflows, entity resolution systems, evaluation harnesses, and the production tooling that keeps it all running. You'll work closely with engineers and data scientists who've shipped real ML systems at scale, and your work will feed directly into decisions made by supply chain teams at some of the country's leading health systems.

Day-to-Day

• Build and maintain components of Clarium's master data enrichment pipeline
• Design and own classification and entity resolution workflows that combine deterministic logic and LLMs for production data processing
• Build and operate evaluation harnesses, label sets, and regression suites (we use Braintrust) to measure and improve pipeline quality with confidence
• Write production Python and SQL; the majority of your time will be spent in code, not in configuration tools
• Analyze complex datasets using statistics and ML to surface actionable insights and inform pipeline improvements
• Proactively audit data for quality issues; find the problems no one else has noticed yet, diagnose root causes, and ship fixes

What You'll Bring

• Strong Python skills and a track record of writing production code, not just scripts or notebooks
• Strong SQL, including complex joins, window functions, performance tuning, and data modeling
• Comfort working in ambiguous environments; you can scope a problem, make a plan, and execute without hand-holding
• A genuine, non-negotiable commitment to data quality; you treat silent bugs as real failures
• Ability to go deep on an unfamiliar domain and develop meaningful expertise over time

Nice to Have

• Experience with LLM integrations, prompt evaluation, or classification at scale
• Familiarity with eval frameworks such as Braintrust, Promptfoo, or equivalent
• Prior work in healthcare, supply chain, or another domain where data quality has direct operational consequences

Skills & Tools

• Need to Know: Python, SQL, PostgreSQL, CI/CD, Production observability
• Nice to Know: Temporal, Braintrust, Snowflake, AWS Sigma