Role overview
As a Senior Applied Machine Learning Engineer on the Detect team, you'll own the end-to-end lifecycle of the AI models that power our camera-based detection system — from data collection and labeling through training, evaluation, and production deployment. Today, Detect uses frontier multi-modal vision models to analyze video streams and detect falls in real time. Your job is to make these models dramatically better and more capable.
You'll take ownership of our ML experimentation platform and infrastructure, maturing it into a robust system that enables the team to rapidly iterate on model quality at scale. You'll design repeatable fine-tuning pipelines that allow models to continuously improve with new production data, and expand the system's detection capabilities beyond falls into new behavioral categories. This is a hands-on, high-autonomy role where you'll directly impact the accuracy of a life-saving system used every day by caregivers across the country.
What you'll work on
- Own and evolve our ML experimentation platform, maturing existing infrastructure into a production-grade system the team relies on daily
- Build data pipelines for collecting, labeling, and preparing production video and image data for model training
- Design repeatable fine-tuning and evaluation pipelines that enable rapid experimentation and measure model performance at scale
- Improve detection accuracy and reduce false positives through prompt engineering, model fine-tuning, and novel inference strategies
- Expand detection capabilities into new behavioral categories
- Work closely with the backend engineering team to integrate model improvements into the real-time video processing pipeline
What we're looking for
- Experience with cloud AI platforms (Google Vertex AI, AWS SageMaker, or similar)
- Experience fine-tuning multi-modal models (VLMs) or large language models (LLMs)
- Familiarity with Kotlin, Java, or similar JVM languages
- Background in computer vision, video processing, or working with image/video data at scale
- Experience building internal ML tooling (labeling, experiment tracking, evaluation)
- Experience maturing early-stage internal tools into production-grade systems
- Full-stack capability with TypeScript/React for building internal tool UIs