Krea
Engineer, Supercomputing & Distributed Systems
San Francisco · On-site · Full-time · Posted Apr 3, 2026
- Visa sponsorship
- Open to sponsoring international visas where possible, including STEM OPT, OPT, H-1B, O-1, and E-3.
About the job
Build and operate distributed systems and GPU infrastructure for Krea’s AI research, training, inference, and data pipelines.
What you'll do
- Design large-scale data pipelines and distributed data systems for processing and annotating multimedia datasets.
- Manage GPU clusters and build systems for distributed training, inference, and job orchestration.
- Profile and optimize training infrastructure, networking, and fault tolerance in collaboration with researchers.
What they ask for
- Strong intuition for distributed systems and how systems behave under different conditions.
Nice to have
- Experience with large-scale ETL, distributed training, or streaming and event-processing systems.
- Knowledge of containerization, operating systems, file systems, and networking.
Tools
Python, Kubernetes, Torch, DuckDB, Arrow, PyTorch, Kafka, InfiniBand
AI focus
Build infrastructure and data pipelines for AI research, model training and inference, dataset preparation, and applied ML.
Model-written summary · the original posting is authoritative