Senior ML Engineer

Truemed

4d•$205,000 - $250,000•Remote

About The Position

Truemed enables health and wellness merchants to accept HSA/FSA funds. Its ML system determines whether products qualify for pretax spending, a classification problem with regulatory, financial, and clinical consequences. The Senior ML Engineer will run experiments to improve the system and bring successful approaches into production, helping Truemed expand its merchant and product-category coverage while maintaining high accuracy. The role involves developing and evaluating classifiers, analyzing errors, building datasets and metrics, testing new models and research, and exploring applications beyond product classification. The engineer will work with clinical operations and legal teams to translate eligibility standards into labeled data and testable system behavior, document results, and help increase AI fluency across engineering. Truemed’s mission is to help people invest in healthy habits and take control of their health. The company supports patients using pretax HSA/FSA funds for healthy food, exercise equipment, and supplements, and is building infrastructure for health and wellness merchants to accept these funds.

Requirements

At least 5 years of engineering or applied research experience, including at least 2 years of hands-on work with LLMs or other ML models on real problems.
A track record of rigorous experimentation, including clear hypotheses, sound baselines, ablations, and defensible conclusions.
Enjoyment and willingness to inspect data directly for extended periods as part of ML work.
Comfort designing evaluation methodology for AI systems, defining correctness, and measuring it.
Hands-on experience with prompting, fine-tuning, retrieval, or agent design on foundation models, including OpenAI, Anthropic, or open-weight models.
Fluency in Python and sufficient engineering depth to take an experiment into production.
Senior-level ownership, including taking an ambiguous ML problem from initial experiment to production independently.
Comfort with ambiguity and discovering the right approach through experimentation.
Public evidence of experimental work, such as a paper, open-source ML work, or a technical write-up of defensible results.

Nice To Haves

Experience with healthcare or regulatory classification.
Experience with multi-model orchestration.

Responsibilities

Run experiments to improve classification accuracy, test hypotheses against labeled data, and ship successful approaches to production.
Design and iterate on product eligibility classifiers using prompting, retrieval, fine-tuning, and multi-model approaches.
Analyze misclassifications to identify failure modes and develop follow-up experiments.
Build and extend evaluation sets and metrics, including methods for labeling ground truth.
Track new models and research and test approaches that could improve accuracy or costs.
Extend classification coverage to new product categories.
Identify and prototype worthwhile applications of Truemed’s models and data beyond product classification.
Collaborate with clinical operations and legal teams to translate eligibility standards into labeled data and testable system behavior.
Document experimental results to support evidence-based model decisions and help raise AI fluency across engineering.