6 months ago
Responsibilities
- Optimize GPU and cluster utilization across thousands of devices for inference, training, data processing, and large-scale deployment.
- Build scalable frameworks for high-volume multimodal data ingestion and processing, distributed model training, and continuous evaluation and benchmarking.
- Develop observability, tracing, and visualization tools for compute clusters to diagnose reliability and performance bottlenecks.
- Collaborate with AI researchers and engineers to integrate acceleration techniques such as custom CUDA kernels and distributed training libraries into production pipelines.
- Manage and optimize Kubernetes and Ray infrastructure for elastic, cost-efficient scaling of distributed systems.
Requirements
- Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent practical experience.
- 5+ years of full-time industry experience in large-scale MLOps, AI infrastructure, or HPC systems.
- Experience with Ray, Apache Spark, and LanceDB.
- Strong proficiency in Python and a high-performance language such as C++.
- Hands-on experience with Kubernetes, Ray, and distributed computing frameworks.
- Experience with PyTorch, TensorFlow, or JAX.
- A master's or PhD in Computer Science or a related technical field is preferred.
- Preferred experience driving projects from conceptual design through production deployment across cross-functional teams.
- Preferred experience building infrastructure for generative AI models, including diffusion models, GANs, or large language models.
- Preferred experience operating large-scale data infrastructure for petabytes of multimodal video, audio, and text data.
- Expertise in GPU acceleration and low-level compute programming, including CUDA and NCCL, is preferred.
Benefits
- Competitive salary and benefits package
- Dynamic and inclusive work environment
- Professional growth and advancement opportunities
- Collaborative culture valuing innovation and creativity
- Access to the latest technologies and tools
Tech Stack
Categories
About HeyGen
HeyGen is a message-first AI video platform that helps people and AI agents turn ideas into professional video in minutes. Used by a community of over 30 million users and 85% of the Fortune 100, HeyGen makes it easy to create and share videos seamlessly across languages and formats.