Senior Software Engineer at Tavus

You will own the end-to-end GPU and cloud infrastructure powering real-time AI agents. By provisioning clusters, managing Kubernetes environments, and optimizing inference workloads, you ensure the reliability and scalability of Tavus’s core compute foundation. This is a high-impact, hands-on role working at the intersection of machine learning, real-time video, and distributed systems.

• Architect and scale GPU infrastructure across cloud providers like AWS, CoreWeave, and Lambda Labs to support training and real-time inference.
• Design and maintain production-grade Kubernetes (EKS) clusters, ensuring high availability, uptime, and efficient resource orchestration for model serving.
• Collaborate cross-functionally with research and product teams to optimize developer experience and internal platforms for deploying cutting-edge ML models.

• Extensive hands-on experience operating and provisioning GPU clusters in production environments, moving beyond simple resource consumption to infrastructure ownership.
• Deep proficiency in Kubernetes and AWS ecosystems, with a track record of scaling distributed systems and managing multi-cloud or hybrid GPU strategies.
• A startup-ready mindset focused on ownership and reliability, ideally with experience in AI-native environments or high-performance machine learning infrastructure.

• Work at a Sequoia and CRV-backed Series B startup with $64M in funding, tackling the most complex compute challenges in generative video AI.
• Own the mission-critical GPU infrastructure that serves as the foundation for real-time multimodal AI agents used by enterprise customers.
• Join a high-caliber engineering team of 50 where you have high autonomy to build and scale infrastructure from scratch.