Senior Applied AI Engineer

The Opportunity

Abby Care is building an AI-native operating system for home care—one that manages the full lifecycle from caregiver and patient intake through care delivery, billing, and compliance. As we expand our capacity to support more family caregivers at scale, we’re looking for a Senior Applied AI Engineer to design and build production AI systems that automate complex, consequential healthcare workflows.

Reporting to the Applied AI Leader, you will own major initiatives from initial problem definition through production deployment and measurable impact. Depending on the problem, you may build an agent that collects clinical documentation, interprets evidence across multiple records, prepares a prior authorization packet, assists a nurse during a visit, identifies billing risk, or coordinates operational follow-up. You will determine how workflows should be decomposed, what context and tools an AI system needs, how its behavior should be evaluated, where deterministic controls are required, and when a human must intervene.

You will work closely with product engineers, product managers, designers, clinicians, operators, and analysts to ensure that the systems you build reflect real-world workflows and produce meaningful outcomes. High-agency and pragmatic, you are comfortable operating in ambiguity, moving between experimentation and production engineering, and owning technically complex systems end to end.

This is a full-time hybrid opportunity based in San Francisco, California, with four days per week in person.

What you’ll work on

• Own high-impact AI systems end to end: Lead initiatives from workflow discovery and technical design through production deployment, iteration, and measurable business or care outcomes.
• Build reliable agents and copilots: Develop systems that reason over structured and unstructured data, use internal tools, and execute multi-step workflows across intake, care delivery, billing, and operations.
• Engineer for real-world complexity: Design systems that handle incomplete information, conflicting evidence, changing requirements, and operational edge cases by combining models, retrieval, rules, state machines, and human oversight.
• Build rigorous evaluation and monitoring: Create representative datasets, evaluation harnesses, production tracing, confidence-based escalation, and systematic error-analysis processes.
• Integrate AI into complete workflows: Partner with Product, Design, Clinical, Operations, Analytics, Platform, and System of Record engineers to ensure AI capabilities solve real problems rather than remain isolated prototypes.
• Raise the technical bar: Contribute reusable platform capabilities, shape architectural decisions, mentor engineers, and improve how the team builds and operates Applied AI systems.

What success looks like

• You independently own and deliver complex Applied AI systems from discovery through sustained production operation.
• The systems you build create measurable improvements in workflow completion, speed, quality, operational capacity, or family experience.
• AI behavior is supported by representative evaluations, systematic error analysis, and clear production monitoring.
• Your systems respond safely to uncertainty through confidence handling, escalation, and human approval where appropriate.
• AI capabilities are integrated into end-to-end product and operational workflows rather than remaining isolated prototypes.
• The components you build are reusable across use cases and improve the velocity of the broader engineering organization.
• You help other engineers make better architectural and implementation decisions through technical leadership and mentorship.
• You develop a deep understanding of Abby Care’s operational and clinical domains and use that understanding to improve the systems you build.

What you’ll have

• 5+ years of professional software engineering, machine learning, or Applied AI experience.
• Experience building and operating AI- or machine-learning-powered systems in production.
• Strong hands-on experience with language models, retrieval systems, structured generation, model APIs, or agent architectures.
• Strong software engineering fundamentals, including backend systems, APIs, data modeling, testing, and observability.
• Experience designing evaluations and analyzing the behavior of probabilistic systems.
• Demonstrated ability to take ownership of ambiguous, cross-functional problems and drive them through implementation and iteration.
• Strong judgment about where AI is appropriate and where deterministic approaches or human review are required.
• Ability to communicate complex technical concepts clearly to technical and nontechnical partners.
• Experience mentoring engineers, leading technically complex projects, or influencing architectural decisions.
• Must reside in or be willing to permanently relocate to the San Francisco Bay Area.
• Experience with model fine-tuning, post-training, reinforcement learning, synthetic data generation, multimodal models, document intelligence, knowledge graphs, ontologies, workflow engines, or state machines is preferred but not required.
• Experience in healthcare, fintech, insurance, or another regulated and operations-intensive environment is preferred but not required.