AI Engineer

Job Description

Job Description

Description:

About Us

AAIT Health (Advanced Artificial Intelligence Technology Health) is building a modern, HIPAA-compliant Electronic Medical Records (EMR) platform. We’re focused on turning today’s best AI and LLM capabilities into reliable, secure, production-grade workflows embedded directly into the EMR experience, so clinicians and staff can work smarter, faster, and with less administrative burden.

What You’ll Do

• Build agentic AI systems that can execute multi-step workflows (e.g., chart review ? summarize ? recommend next actions ? draft documentation ? route tasks) with appropriate human oversight.

• Design and implement tool-using LLM workflows (function calling / tools, retrieval, structured outputs, planner–executor patterns, and guardrails).

• Integrate AI capabilities into our EMR via backend services and APIs (e.g., .NET Core services, MySQL, and modern frontend clients).

• Implement retrieval-augmented generation (RAG) and clinical knowledge workflows that query patient context, incorporate medical reference content, and cite sources with traceability.

• Engineer safety, privacy, and compliance into AI workflows, including PHI-safe processing, audit logs, role-based access, minimum-necessary data, prompt/data redaction, and secure storage.

• Evaluate and improve quality using automated and human-in-the-loop evaluation (e.g., grounding, hallucination rates, task success, latency, and cost).

• Deploy and operate AI services in production, including monitoring, rate limiting, fallbacks, caching, incident response, and model/provider switching.

• Collaborate cross-functionally with product, clinical SMEs, security/compliance, and engineering to ship AI-powered features that users trust.

Role Details

• Full-time position with presence in the office required.

• Core schedule: Monday–Thursday 8:00 a.m. to 5:00 p.m. and Friday 8:00 a.m. to 12:00 p.m., with occasional work outside regular business hours as needed.

• Travel may be required.

• The position operates in a professional office environment and involves significant time writing, typing, speaking, listening, standing, sitting, walking, and reaching.

• Operation of standard office equipment, non-CDL motor vehicles, mobile phones, and related technology is expected.

Requirements:

You Might Be a Fit If

• You have strong software engineering fundamentals and production experience (APIs, testing, debugging, performance).

• You have hands-on experience building with LLMs (OpenAI/Anthropic/others), including tool/function calling, structured outputs, and retrieval-augmented generation (RAG).

• You’ve integrated AI into real products (not just notebooks or demos) and understand the tradeoffs of latency, cost, and quality.

• You can design for reliability with deterministic interfaces, schema validation, retries, fallbacks, and evaluation.

• You are comfortable working across backend and data, and optionally some frontend integration.

• You communicate clearly, enjoy collaborating with cross-functional partners, and can explain complex AI behavior to non-technical stakeholders.

• You are comfortable operating in a startup-like environment, prioritizing impact and iterating quickly while maintaining quality.

Bonus Points

• Experience in healthcare, EMR, or clinical workflows; familiarity with standards such as HL7/FHIR.

• Experience with .NET Core, MySQL, and cloud deployment (Azure is a plus).

• Familiarity with security and compliance practices such as HIPAA and SOC2-style controls, including RBAC and audit logging.

• Experience building evaluation pipelines (golden datasets, offline eval, red-teaming, prompt regression tests).

• Experience with vector databases/search or building retrieval layers over relational and document stores.

• Knowledge of PHI de-identification, redaction, and safe data handling.

Tech Stack

• Frontend: React + Vite + TypeScript

• Backend: .NET Core services

• Database: MySQL

• Cloud: Azure, AWS

• AI: LLM APIs, retrieval/vector search, observability & evaluation tooling