Knowledge Routing Engineer

Sage Care Inc • Palo Alto, CA

4d

About The Position

Sage Care is a fast-growing, early-stage healthcare startup transforming healthcare by simplifying care navigation. Our platform makes it easier for patients to find the right doctor and helps providers focus on those who need them most through harnessing the latest AI innovations. We have expanded internationally to the MENA region and are partnering with health systems there to deploy our AI-powered care navigation platform.

Every day, our services match real patient queries, symptoms, and pathologies to providers, sites of care, and urgency. These mappings are highly complex and non-linear. Currently, much of this mapping is hand-tuned and heuristic driven. We are looking for someone to own the knowledge encoding piece of this puzzle, helping us learn from real patient data and medical diagnoses and symptoms to build out an encoded representation that can translate real patient requests into actionable results. This role sits at the intersection of ML/AI research and software engineering. You will help build the core routing engine for our agent, working closely with engineers who work on matching software, and medical experts to guide an informed, encoded knowledge representation.

Requirements

• 7+ years of ML engineering experience
• Experience working with foundational ML models (e.g. learned decision trees, deep learning, and reinforcement learning)
• Strong backend engineering skills and systems thinking
• Experience working with ambiguous problems and defining solutions from first principles
• Experience turning research or novel techniques into testable prototypes

Nice To Haves

• Experience with evaluation frameworks and model quality measurement
• Experience with medical AI systems
• Experience designing human-in-the-loop workflows for machine learning

Responsibilities

• Build and own the core symptom and query routing engine
• Build models that can map complex queries and symptoms to urgency classifiers, providers, and specialties
• Build learned decision trees that can infer if there are necessary follow up questions to ask to gain more insight into the patient’s specific query
• Leverage insights and learning from real protocols (e.g. Schmidt-Thompson) as well as other triaging SOP’s
• Work with medical professionals to build generalizable representations of how queries and symptoms can map to body systems, specializations, and restrictions
• Learn from real data
• Build self-learning models that can learn and iterate from real user data and diagnoses
• Establish metrics of quality and hill climb on these to improve the model in the long term

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed