Forward Deployed ML Engineer (AI)

Location: New York, New York, United States; On-Site

Salary: USD 170,000–190,000 per year

Company

hirify.global develops AI systems that extract structured oncology data from unstructured clinical documents for pharmaceutical companies and cancer hospitals.

What you will do

• Design and build agentic extraction pipelines for 500+ page patient charts, including clinical notes, pathology reports, imaging reports, and genomic panels.
• Own extraction accuracy by creating evaluation datasets, analyzing precision and recall per variable, identifying failure modes, and improving agent architectures.
• Read and interpret clinical documentation to improve extraction of ambiguous oncology data such as disease stage, biomarker status, and line of therapy.
• Collaborate with clinical annotators to create gold-standard datasets and resolve edge cases.
• Coordinate with customer data science and clinical teams to clarify data dictionaries and close quality gaps.
• Deploy, scale, monitor, and improve production pipelines with internal engineering and infrastructure teams while meeting customer deadlines.

Requirements

• 2+ years of experience building ML or AI systems in production.
• Experience building and deploying AI agents or multi-step LLM pipelines, including agent architecture, tool use, and orchestration frameworks.
• Strong Python skills for pipeline development, data processing, and infrastructure integration.
• Practical experience with prompt engineering, fine-tuning, RAG, and LLM evaluation design.
• Experience building evaluation frameworks for LLM-based document extraction, including precision, recall, per-class analysis, and error taxonomies.
• Willingness to become a domain expert in oncology data and communicate directly with customer clinical and data science teams.

Nice to have

• Knowledge of agentic ML frameworks, production patterns, and failure modes.
• Clinical or biomedical NLP experience.

Culture & Benefits

• Work in a customer-facing role with ownership across the full delivery cycle.
• Operate across clinical, data science, engineering, and infrastructure teams.
• Work in intensive delivery sprints followed by iteration and improvement cycles.
• Build production systems for pharmaceutical and cancer hospital use cases.