Machine Learning Engineer

Overview In this role you will design and deploy autonomous AI agents that triage patient messages and manage clinical workflows within ModMed’s patient engagement platform. You will build conversational interfaces, enable tool use with external systems, and continuously evaluate models for safety, accuracy, and reliability. The work directly supports safer, more efficient patient care while aligning with HIPAA requirements and clinical standards. You’ll collaborate with cross-functional teams to push the boundaries of AI in healthcare and drive measurable impact. Compensation / BenefitsComprehensive medical, dental, and vision benefits with HSA401(k) match on up to 6% of compensationGenerous PTO and Paid Parental LeaveLife and Disability benefitsFlexible Spending Account and employee assistance programsProfessional development including tuition reimbursement and LinkedIn Learning ResponsibilitiesDesign and deploy multi-agent systems to execute complex clinical workflowsBuild conversational AI interfaces for message triage, sentiment analysis, and context-aware responsesCreate tool-calling architectures enabling AI agents to interact with EHRs, databases, and external APIsDevelop evaluation frameworks to measure accuracy, reliability, safety, and task completionImplement guardrails and safety measures to ensure HIPAA-compliant, trustworthy outputsCreate AI observability dashboards tracking reasoning traces, performance, and production task completion Key requirements3+ years in Machine Learning with focus on AI agents and conversational NLPExperience with LangGraph, AutoGen, CrewAI or similar no-code/low-code frameworksHands-on model fine-tuning and rigorous evaluation methodologiesFamiliarity with detecting hallucinations, bias, and failure modes in agent systemsStrong knowledge of guardrails and safety mechanisms for sensitive environmentsHands-on AWS and Databricks experience, including CI/CD for ML deploymentsProficiency in Python and experience with evaluation frameworks like RAGAS or DeepEvalcollaboration with cross-functional teamsproblem solving and critical thinkingattention to safety and complianceLangGraphAutoGenCrewAI