Kernel Engineer New Grad
Cerebras Systems, Inc.Visit Cerebras Systems, Inc. website
Cerebras builds wafer-scale AI computing systems and a cloud inference platform for training, fine-tuning, and serving AI models.
Sunnyvale, California, United States
About Cerebras Systems, Inc.
Cerebras Systems is an AI-infrastructure company founded in 2015. It sells rack-scale wafer-scale computing systems and provides cloud-based, API-accessible AI inference alongside on-premises deployments.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will design, implement, optimize, debug, and validate high-performance machine learning and linear algebra kernels. You will apply parallel algorithms, profile performance, investigate correctness and utilization issues, build tests, and collaborate on low-level software for specialized hardware.
Requirements
- C++
- Python
- Computer architecture
- Processor
- Memory hierarchy
- Instruction execution
- Data movement
- Data structure
- Algorithm
- Software development
- Debugging
- Parallel computing
- Performance optimization
- Hardware-software co-design
- Kernel development
- Compiler
- HPC
- Systems programming
- Parallel algorithm
- Multithreading
- Distributed memory
- GPU
- FPGA
- Assembly language
- CUDA
- OpenCL
- Machine learning
- Neural network
- PyTorch
- TensorFlow
- Numerical computing
- Linear algebra
- Profiling
- Benchmarking
- Performance analysis
- API development
Responsibilities
- Design and implement machine learning and linear algebra kernels
- Develop and debug high-performance kernel routines using low-level programming
- Apply parallel algorithms to map workloads onto the architecture
- Analyze performance data and profiling results to guide design decisions
- Investigate correctness, performance, and hardware-utilization issues
- Develop unit tests and system-level validation methodologies
- Collaborate to improve software and system performance
- Study emerging machine learning workloads and evolve kernel libraries
- Participate in code reviews and technical discussions
- Build knowledge of processor architecture, instruction sets, memory systems, and communication models