Healthcare AI Scientist — LLM Agents & EHR Data

Verily Health; Mountain View (CA); On-site; USD 219,000 - 246,000; Full time

Job summary

Verily Health in Mountain View, CA, seeks an AI Applied Scientist to advance data curation, healthcare AI, and multi‑agent systems. You will design scalable pipelines for EHR and claims data, and work on cutting‑edge LLMs and intelligent agent architectures to extract clinical meaning from messy text.

You will implement robust evaluation frameworks to ensure safety, factuality, and regulatory compliance, while collaborating with cross‑functional teams across clinical and technical domains.

Qualifications

• Master’s degree in a quantitative discipline or equivalent practical experience.
• Minimum of 3 years of industry experience applying ML/NLP and AI techniques to healthcare data.
• Direct experience curating real‑world data such as EHRs or medical claims.
• Strong proficiency in Python and standard ML libraries (PyTorch, TensorFlow).

Responsibilities

• Design, develop, and deploy generative AI workflows and multi‑agent architectures for personalized interactions.
• Enhance NLP tools to automate abstraction of clinical variables from unstructured text.
• Create automated evaluation pipelines to benchmark safety, accuracy, and compliance across models.
• Handle real‑world data challenges with privacy preservation and rigorous validation before production.
• Present technical results and evaluation methods to both technical and non‑technical audiences.

Skills

Python

ML/NLP proficiency

Real-world data curation

Generative AI

Education

Master’s degree in a quantitative discipline

PhD degree (preferred)

Tools

PyTorch

TensorFlow

Job description

Verily Health in Mountain View, CA, seeks an AI Applied Scientist to advance data curation, healthcare AI, and multi‑agent systems. You will design scalable pipelines for EHR and claims data, and work on cutting‑edge LLMs and intelligent agent architectures to extract clinical meaning from messy text.

You will implement robust evaluation frameworks to ensure safety, factuality, and regulatory compliance, while collaborating with cross‑functional teams across clinical and technical domains.