ML Infra/Systems Engineer

We’re partnering with an early-stage AI startup building cutting-edge visual conversational systems that allow users to interact with AI over real-time video. Their platform combines multimodal ML, real-time streaming, and scalable infrastructure to deliver human-like AI experiences. They are now hiring an ML Infrastructure / Systems Engineer to build and optimize the foundational infrastructure powering these multimodal AI workloads.

This is a hands-on role with significant ownership, focused on distributed systems, GPU clusters, video streaming, and large-scale ML pipelines.

Key Responsibilities:

• Design, build, and optimize the serving stack for multimodal AI workloads, focusing on latency, throughput, and cost.

• Architect and maintain infrastructure for real-time WebRTC connections to ensure smooth video and audio streaming.

• Build and orchestrate robust, distributed data pipelines for offline processing, evaluation, and model training using frameworks such as Dagster or Ray.

• Configure, maintain, and optimize GPU clusters and other compute infrastructure using Kubernetes, Terraform, and cloud platforms.

• Develop CI/CD, model evaluation, and versioning systems to support safe, zero-downtime deployments and rapid iteration.

• Collaborate closely with ML researchers and product engineers to design scalable, reliable systems that power visual conversational AI.

Key Qualifications:

• 2+ years of experience building and operating ML infrastructure in production environments.

• Strong experience designing and running distributed data pipelines.

• Hands-on experience with production reliability, monitoring, incident response, and capacity planning for high-traffic services.

• Practical experience with video, audio, or other multimedia ML workloads.

• Proficiency in Python and either Rust or Go.

• Strong experience with Kubernetes, Terraform, and cloud platforms.

• Experience in VC-backed startups or high-growth technology companies is a plus.

• Comfortable working on-site in Seattle 5 days a week.

• Hands-on engineering focus; candidates who are purely in management or director-level roles without recent coding experience are not a fit.

• Candidates should have a broad understanding of ML infrastructure and not be limited to one set of technologies.