Meshy
AI Infrastructure Engineer
About this role
Meshy, a leading 3D generative AI company, seeks an AI Infrastructure Engineer to design and optimize the inference platform that powers their model serving stack. You'll work on GPU resource management, service orchestration, and production reliability while supporting rapid growth in a well-funded startup.
What you'll do
- Design and optimize core inference platform capabilities including services, task scheduling, orchestration, and elastic scaling
- Develop CPU/GPU resource management systems to balance stability, utilization, and cost efficiency across inference and training workloads
- Implement unified GPU resource scheduling and explore technologies like MIG, MPS, and virtualization in production
- Optimize throughput, latency, and availability across complex inference pipelines and high-concurrency scenarios
- Drive R&D efficiency, cost management, and disaster recovery architecture to support company scaling
- Research AI-native infrastructure and automated operations to improve system reliability and usability
What they're looking for
- Go or Python programming with strong software engineering practices
- Kubernetes, Docker, and container orchestration
- Distributed systems and microservices architecture
- Linux, operating systems, computer networks fundamentals
- GPU inference platforms and resource scheduling
- CI/CD, build systems, and deployment infrastructure
- Model serving and task orchestration frameworks
- Problem-solving and debugging complex production systems
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Meshy
Meshy builds a 3D generative AI platform that transforms 3D content creation through advanced AI systems and high-performance graphics technology. The company is hiring infrastructure engineers, fullstack engineers, and graphics specialists to scale its GPU-intensive platform, optimize inference systems, and build production-grade rendering pipelines.
- Website
- meshy.ai
Likely interview questions
- Walk us through a production incident where you had to optimize latency or throughput in an inference pipeline—what was the bottleneck and how did you resolve it?
- Describe your experience with GPU resource management and scheduling. Have you worked with constraints like sharing GPUs across multiple inference or training workloads?