Omada Health Scales Patient Care with Fine-Tuned Llama Models on Amazon SageMaker

Omada Health Scales Patient Care with Fine-Tuned Llama Models on Amazon SageMaker

Omada Health, a virtual healthcare delivery company, has launched a new nutrition experience in 2025 featuring OmadaSpark, according to an AWS blog post co-written with Omada Health’s AI/ML Product Manager Sunaina Kavi.

OmadaSpark is described as an AI agent that was trained with clinical input and delivers real-time motivational support to patients. The system was built by fine-tuning Llama models on Amazon SageMaker AI, AWS’s machine learning platform.

According to the source material, the implementation on SageMaker enabled Omada Health to scale their patient care capabilities. The AI agent focuses on nutrition guidance, representing what the post describes as an innovation in the company’s virtual healthcare offerings.

The article is co-authored by AWS and Omada Health representatives, indicating a collaborative case study showcasing how healthcare organizations can leverage AWS’s AI infrastructure for clinical applications. However, specific details about performance metrics, patient outcomes, or technical implementation specifics were not included in the provided excerpt.