Full-Stack AI Engineer

Liza Health•Cambridge, MA

3d•Hybrid

About The Position

Liza Health is a Cambridge-based startup building an AI-enabled platform for modern primary care. It works alongside primary care practices to reduce administrative burden, support stronger patient relationships, and help patients receive appropriate care. The company was co-founded by physician Dr. Rushika Fernandopulle, entrepreneur and investor Andy Palmer, and healthcare entrepreneur Duncan Reece. Liza is hiring AI Engineers to design, build, and ship AI systems for its care model. This generalist role owns problems end to end, working directly with clinicians and care teams to develop healthcare applications for patients, clinicians, and clinical operators. Work may span prompts and evaluation suites, EHR integrations, frontend features, and healthcare data pipelines. The engineering approach includes testing agent changes, using LLM judges to score production replies, and reviewing production traces. The role is based in Cambridge, Massachusetts, with in-person collaboration. Hybrid work is encouraged; most people are in the office two to three days per week, with flexibility especially after ramp-up. The company offers a competitive salary, meaningful equity, and comprehensive health, dental, and vision benefits.

Requirements

At least 5 years of software engineering experience; the company’s stack is TypeScript end to end.
Hands-on experience shipping LLM-powered products to real users at scale.
Experience designing, testing, and optimizing prompts, tool descriptions, and retrieved context, and understanding frontier models’ strengths and failure modes.
Experience building agentic workflows with function calling, multi-step planning, memory, state management, and protocols such as MCP; ability to build multi-agent systems with human oversight.
Experience building evaluation datasets and rubrics with domain experts, LLM-as-a-judge and automated graders, human-rating workflows, and regression detection; ability to analyze transcripts and improve quality measurably.
Experience with RAG pipelines, embeddings, structured knowledge, and memory for reasoning over large, heterogeneous data.
Full-stack and systems breadth across APIs, data infrastructure, web and mobile frontends, cloud deployment, CI/CD, containers, and production observability; ability to take a system from prototype to reliable production.
Fluency with coding agents and AI-assisted development workflows.
Ability to work as a generalist, learn new tools and domains quickly, and handle backend services, data pipelines, frontends, prompts, and evaluations.
Strong communication and collaboration skills in small, cross-functional teams; comfort giving and receiving direct feedback and sharing context.
Ability to frame ambiguous problems, make decisions with incomplete information, and own outcomes.
Mission-driven interest in improving U.S. healthcare and primary care.
No prior healthcare experience or clinical credentials are required.

Nice To Haves

Experience with EHR and interoperability technologies such as Epic, Oracle Health (Cerner), athenahealth, HL7, FHIR, SMART, health information exchanges, or data aggregators such as Metriport, Zus, Gorilla Health, and Redox.
Exposure to primary care, value-based care, care coordination, or patient access workflows.
Experience fine-tuning or post-training models, or building supervised fine-tuning and reinforcement-learning data programs.
Voice or multimodal AI experience, such as patient-facing phone agents or ambient documentation.
Prior work as a forward-deployed engineer, technical founder, or early engineer at a startup.
Open-source contributions, publications, or side projects that demonstrate engineering craft.
Startup experience.

Responsibilities

Build production AI applications and agentic systems for patients, clinicians, and care teams across scheduling, intake, documentation, care coordination, outreach, and clinical decision support.
Partner with patients, physicians, health coaches, and operations staff to translate real-world workflows into technical requirements and measurable outcomes.
Work with the AI-native design team to create simple, efficient, and effective user experiences.
Design, test, and iterate on prompts, context, and tool definitions; keep them versioned, reviewed, and safe to change.
Define and run evaluations, including building clinician-informed datasets and rubrics, implementing automated graders, reviewing transcripts, and setting launch criteria.
Integrate AI with healthcare systems, including the FHIR backend (Medplum), practice EHRs, health information exchange aggregators, HL7 interfaces, claims and scheduling systems, and the data platform.
Build with PHI, HIPAA, privacy, security, and auditability in mind, including human-review workflows and clinical-risk escalation paths.
Monitor production systems and use error analysis, observability, and user feedback to improve reliability, cost, latency, and model selection.
Develop reusable building blocks such as evaluation harnesses, agent patterns, integration libraries, and internal tools.
Work across the TypeScript monorepo, including APIs, web and mobile apps, infrastructure, and agents.