Tech - Staff AI Engineer, Agent Platform
About The Position
Onpoint Healthcare Partners builds Iris, a Medical Agent AI platform that takes administrative work off the plates of providers and care teams. Our agents handle charting, coding, care gap closure, and care coordination across the full patient journey, combining AI automation with expert clinical oversight. More than 2,000 providers across 35+ specialties rely on the platform every day, which means the systems behind it have to be accurate, reliable, and auditable without exception. We are seeking a Staff AI Engineer to design, build, and scale the agent systems behind Iris. This role combines strong software engineering fundamentals with deep experience building production AI systems. The work you ship directly determines how much time providers get back for patient care. The ideal candidate has shipped AI solutions to production that deliver measurable business value, and will evolve and be a good steward of the agent platform architecture.
Requirements
- 8+ years of software engineering experience.
- 3+ years building AI and ML applications.
- Strong Python and/or C# development experience.
- Experience deploying AI systems into production environments.
- Experience with LLMs, RAG architectures, agent frameworks, and AI evaluation.
- Experience with AWS and distributed systems.
- Hands-on experience building MLOps infrastructure such as training pipelines, model registries, feature stores, A/B testing, and drift detection.
- Strong debugging, observability, and operational skills.
Nice To Haves
- Healthcare industry experience.
- Experience with HIPAA, PHI, and regulated environments.
- Experience with vector databases, embeddings, and knowledge graphs.
- Experience building AI systems at large scale.
Responsibilities
- Design and develop AI applications and agent workflows.
- Build scalable backend services, APIs, and integrations supporting AI solutions.
- Evolve and steward the agent platform architecture, including orchestration, runtime safety, and prompt governance.
- Treat prompts and tool schemas as versioned code with staged rollouts and rollback.
- Develop and maintain evaluation frameworks for AI agents and models, with CI gates that block bad changes before release.
- Design retrieval, memory, and context management strategies.
- Build observability, monitoring, and debugging capabilities, including full trace and replay for incident resolution.
- Build the MLOps foundation the team is currently missing: training and retraining pipelines, model versioning, model registry, and feature stores, so model work stops being reactive.
- Stand up A/B testing infrastructure and automated drift detection that triggers retraining, so model quality is monitored and maintained without manual firefighting.
- Track token and tool costs per run, workflow, and tenant, and keep costs predictable as usage scales.
- Improve reliability, performance, safety, and cost efficiency of production AI systems.
- Partner with Product, Clinical, Data Science, and Engineering teams to deliver AI capabilities.
- Own AI deployment, monitoring, and operational excellence.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
Education Level
No Education Listed
