Member of Technical Staff, Training Infra
The Inception Company8 months ago
San Mateo, CA, USASenior
Responsibilities
- Design, implement, and optimize distributed training systems that scale across thousands of GPUs and nodes.
- Develop high-performance optimizations to maximize training throughput and efficiency.
- Build reusable frameworks and libraries that improve training reproducibility, reliability, and scalability for new model architectures.
- Debug and maintain performant, maintainable code in complex codebases.
Requirements
- BS, MS, or PhD in Computer Science, Engineering, or a related field, or equivalent experience.
- Understanding of PyTorch and TensorFlow from a systems perspective.
- Strong engineering skills with the ability to contribute performant, maintainable code and debug complex codebases.
- Proficiency in Python and at least one systems programming language: C++, Rust, or Go.
- Experience with Docker, Kubernetes, and CI/CD pipelines.
- Preferred: experience building and maintaining language models with tens of billions of parameters or more.
- Preferred: experience with Kubeflow or Airflow.
- Preferred: background in performance optimization and profiling of ML systems using Prometheus, Grafana, or OpenTelemetry.
- Preferred: familiarity with PyTorch/XLA, DeepSpeed, or Megatron-LM.
Categories
About The Inception Company
The Inception Company provides technology-enabled, fully managed virtual and live event production for life sciences, including webcasts, broadcast, and video. It operates studios in Fairfield, NJ and Ambler, PA, and offers platforms such as the PandoGo meeting experience and The Inception Portal for AV planning. Founded in 2005 and privately held, the company is headquartered in Fairfield and received private equity investment in 2021.