Applied Scientist

Evidium, Inc.; San Francisco (CA); Hybrid; USD 140,000 - 190,000; Full time

Benefits offered by this job

Competitive compensation

Equity

Interdisciplinary team

Job summary

Evidium, Inc. is seeking a Machine Learning Engineer to develop predictive models from longitudinal healthcare data and deploy scalable ML pipelines. You will balance rigor, interpretability, and scalability while collaborating with clinicians, engineers, and product teams.

Join a collaborative, curious culture focused on transparent, clinically grounded AI. Opportunities to publish, shape research direction, and influence policy decisions are available.

Qualifications

• Strong background in statistical modeling or data science with longitudinal data.
• Proficiency in Python and ML ecosystems (PyTorch, JAX, NumPyro, PyMC).
• Experience taking models from research prototype to production.
• Bonus: clinical data terminologies and actuarial modeling.

Responsibilities

• Develop predictive models forecasting disease progression, utilization, and cost from temporal clinical data.
• Design interpretable ML solutions trusted by clinicians and actuaries.
• Research and prototype novel approaches using classical and modern ML.
• Implement robust, scalable ML pipelines for training, validation, and deployment in distributed compute environments.
• Collaborate with data engineers, clinicians, and product teams to align models with real-world healthcare needs.
• Communicate findings and methodologies clearly through visualizations, documentation, and technical presentations.
• EDA, working closely with domain experts.

Skills

Python

Temporal data

Production deployment

Statistical modeling

Survival analysis

Probabilistic methods

Tools

PyTorch

JAX

NumPyro

PyMC

Job description

Machine Learning Engineers who thrive at the intersection of data science, modeling, and software engineering.

Evidium’s mission is to scale medical knowledge and create reliable AI for the health benefit of every person.

We’re developing a new class of neurosymbolic AI systems, combining statistical learning with structured medical knowledge to create transparent, clinically grounded models. Our early adopters include health systems and insurers using Evidium’s technology to improve patient-level foresight, evidence deployment, and prediction.

Our culture is collaborative, curious, and first-principles-driven. We prize individuals who are not only excellent implementers, but original thinkers — people who can reason clearly about complex systems and translate that understanding into production-quality solutions.

The Role

We’re looking for Machine Learning Engineers who thrive at the intersection of data science, modeling, and software engineering. You’ll design and implement models that learn from longitudinal healthcare data, balancing rigor, interpretability, and scalability.

This is a rare opportunity to work on foundational modeling challenges in healthcare, where your work will directly inform clinical, actuarial, and policy decisions.

Responsibilities

• Develop predictive models that forecast disease progression, utilization, and cost from temporal clinical data (claims, EHR, lab, pharmacy, etc.)
• Design interpretable and explainable ML solutions that can be trusted by clinicians, actuaries, and decision-makers
• Research and prototype novel approaches, leveraging both classical and modern ML
• Implement robust, scalable ML pipelines for training, validation, and deployment in distributed compute environments
• Collaborate cross-functionally with data engineers, clinicians, and product teams to align models with real-world healthcare needs
• Communicate findings and methodologies clearly through visualizations, documentation, and technical presentations
• EDA, working closely with domain experts

Requirements

• Strong background in statistical modeling, machine learning, or data science, ideally with experience in temporal or longitudinal data
• Proficiency in Python and ML ecosystems (PyTorch, JAX, NumPyro, PyMC, etc.)
• Demonstrated ability to take models from research prototype to production
• Familiarity with probabilistic methods, survival analysis, or Bayesian inference is a plus
• (Bonus) Experience with clinical data and terminologies (ICD, CPT, SNOMED CT, LOINC, etc.)
• (Bonus) Exposure to actuarial modeling, claims forecasting, or risk adjustment methods

What We Offer

• Competitive compensation and equity
• Work alongside an interdisciplinary team of researchers, engineers, and clinicians
• Direct impact on the development of healthcare’s next-generation predictive infrastructure
• Opportunities to publish, co-author patents, and shape the research direction of Evidium