Job Description
As a Platform Engineer, you'll build and own the software that takes our ML models from research to production: the Theia SDK that runs real-time inference at the edge, the APIs and services our customers and partners integrate with, and the cloud infrastructure behind our data and model pipelines. You'll work closely with our ML engineers and researchers, so you'll need to understand how models are trained, evaluated, and deployed, but your core focus is building reliable, well-designed systems around them.What You'll DoSDK Development: Design, build, and maintain the Theia SDK in Python and C, delivering real-time eye-tracking inference across desktop, mobile, embedded, and XR platforms. Own packaging, performance, versioning, and the developer experience.API & Backend Development: Design and implement APIs and backend services for data ingestion, model serving, licensing, and customer integrations, with a focus on clean interfaces, reliability, and security.Cloud Engineering: Build and operate infrastructure on AWS, including containerized services on Kubernetes, infrastructure-as-code, CI/CD, and monitoring.Full Stack: Build internal tools, dashboards, and customer-facing web interfaces that make our data and models accessible.Data Science Engineering: Build and scale the pipelines that process millions of eye-tracking records for training, evaluation, and analytics. Clean, transform, validate, and analyze large tabular datasets, and build the tooling that lets our ML team explore and work with data efficiently.ML Productionization: Partner with ML engineers to deploy, optimize, and monitor models in production and at the edge, including model conversion, latency and memory optimization, and inference pipelines.
