Website:
martlet.ai
Job details:
About Martlet AI
Martlet AI builds NLP and small language model (SLM) solutions for the Risk Adjustment ecosystem and HCC coding workflows. We're working at the intersection of clinical coding expertise and artificial intelligence, and we need coders who want to shape how AI learns to code, not just code charts.
About the Role
This is not a typical medical coding role. You won't be working a production coding queue. Instead, you'll be the clinical coding brain behind our AI systems — reviewing charts to establish ground truth, building gold-standard datasets, and evaluating how well our models perform against expert human judgment.
Your coding expertise will directly determine how accurate, defensible, and audit-ready our AI becomes. If you've ever looked at a coding tool and thought "I could make this smarter," this role is for you.
What you'll do
- Perform expert-level chart reviews across HCC (risk adjustment) and DRG (inpatient) coding to establish ground-truth annotations for AI training and evaluation
- Create, curate, and maintain gold-standard datasets used to benchmark model performance
- Evaluate AI-generated coding output — validating code capture, MEAT/TAMPER criteria, specificity, and supporting documentation — and systematically log where models get it right and where they miss
- Conduct error analysis with our AI/ML team: categorize model failure patterns, explain the clinical coding rationale behind discrepancies, and help translate coding guidelines into rules and evaluation criteria the models can be measured against
- Stay current on ICD-10-CM guidelines, CMS-HCC model changes (including V28), and OIG/RADV audit trends, and ensure our datasets and evaluation standards reflect them
- Document coding decisions and rationale clearly so they can be used by both auditors and engineers
- Contribute to internal QA standards, annotation guidelines, and coder-AI workflow design
What we're looking for
- Active AAPC certification required (CPC, CRC preferred; CIC/CCS a plus)
- 5+ years of hands-on medical coding experience, with substantial time spent in HCC/risk adjustment coding; DRG/inpatient experience strongly preferred
- Deep working knowledge of ICD-10-CM Official Guidelines, CMS-HCC risk adjustment models, and MEAT criteria
- Experience with chart audits, coding QA, or second-level review
- Meticulous attention to detail — you catch the unsupported code, the missed specificity, the copy-forward problem
- Ability to articulate why a code is right or wrong in writing, clearly and consistently
- Comfort working with spreadsheets, annotation tools, and structured data; curiosity about AI/NLP is a big plus (no engineering background required — we'll teach you the AI side)
- Self-directed and comfortable in a fast-moving startup environment
Nice to Have
- RADV audit or IVA experience
- Experience creating training materials, coding guidelines, or annotation standards
- Prior experience working alongside product or engineering teams
- Familiarity with CDI workflows
Why join us
- Shape the future of AI in risk adjustment — your expertise becomes the standard our models are measured against
- Escape the production-queue grind: this role is about depth, judgment, and quality, not chart counts
- Work directly with the founding team at a focused, fast-growing AI company
- Grow into an emerging career path: AI evaluation and clinical data quality is one of the fastest-growing specialties in health tech
Click on Apply to know more.
