Oncology AI Engineer — End-to-End Agent Pipelines

triomics inc.; New York (NY); On-site; USD 90,000 - 130,000; Full time

Job summary

triomics inc. in New York is seeking a professional to build and deploy AI agent pipelines that extract structured oncology variables from unstructured patient documents. This role involves designing extraction pipelines, ensuring precision and accuracy, and coordinating closely with clinical teams and customers.

The ideal candidate should have 2+ years of experience in ML/AI systems, strong Python skills, and a willingness to dive deep into oncology data. A strong focus on customer deliverables is essential, along with the ability to manage timelines effectively.

Qualifications

• 2+ years building ML/AI systems in production.
• Built and deployed AI agents or multi-step LLM pipelines.
• Strong proficiency in Python for data processing and pipeline development.
• Experience in building evaluation frameworks for document extraction tasks.
• Willingness to become an oncology domain expert.
• Ability to manage high-intensity delivery sprints.

Responsibilities

• Design and build extraction pipelines from patient charts.
• Own accuracy evaluation and improvement.
• Work with clinical annotation teams to build datasets.
• Coordinate with customer and engineering teams for deployment.
• Deliver on customer timelines.

Skills

Building ML/AI systems

Strong Python

Customer-facing communication

Agent architectures

Job description

triomics inc. in New York is seeking a professional to build and deploy AI agent pipelines that extract structured oncology variables from unstructured patient documents. This role involves designing extraction pipelines, ensuring precision and accuracy, and coordinating closely with clinical teams and customers.

The ideal candidate should have 2+ years of experience in ML/AI systems, strong Python skills, and a willingness to dive deep into oncology data. A strong focus on customer deliverables is essential, along with the ability to manage timelines effectively.