An MPI/SHMEM validation engineer plays a crucial role in ensuring the quality and performance of Message Passing Interface (MPI) and SHared MEMory (SHMEM) based applications and middleware, particularly in High-Performance Computing (HPC) environments. They are responsible for testing, debugging, and validating parallel programming frameworks and their implementations to meet established standards and specifications. This involves working with both hardware and software aspects of HPC systems and ensuring optimal functionality and efficiency for communication middleware like MPI and SHMEM.

Key responsibilities

• Test plan development and execution: Designing and executing comprehensive test plans to validate MPI and SHMEM features, functionality, and performance.

• Debugging and Root Cause Analysis: Identifying, analyzing, and resolving issues found during validation and testing, collaborating with development teams to implement corrective actions.

• Performance Evaluation and Optimization: Evaluating and optimizing the performance of MPI and SHMEM based applications and middleware, including communication collective algorithms like AllReduce.

• Automation and Infrastructure Development: Developing and maintaining post-silicon validation infrastructure including software, hardware, and automation environments.

• Collaboration: Working closely with hardware teams, software developers, architects, and various stakeholders to ensure seamless integration and validation of systems.

• Documentation: Generating and maintaining detailed documentation of validation activities, test results, and compliance reports

• Troubleshooting: Providing technical expertise and support for troubleshooting and resolving technical issues related to MPI and SHMEM.

• Staying updated with technology: Maintaining knowledge of validation trends, industry standards, and new technologies in high-performance computing, parallel programming, and communication middleware. 

This position will support government accounts.  Therefore, due to federal export-control regulations, the selected candidate must hold U.S. citizenship

Required skills

• Parallel Programming and Communication: Strong understanding of parallel programming models, specifically MPI and SHMEM, including their concepts, features, and one-sided communication APIs.

• HPC Architectures: Knowledge of high-performance memory subsystems, SoC/ASIC memory architecture, high-speed I/O interfaces, and their interaction with parallel programming models.

• Programming and Scripting: Proficiency in programming languages like C/C++, Python, and potentially others like Perl, for developing validation tests, scripts, and tools.

• Validation Methodologies: Experience with various validation methodologies, including formal analysis and runtime instrumentation, for detecting MPI bugs and ensuring correctness.

• Debugging Tools and Techniques: Expertise in utilizing debugging tools, methodologies, and techniques for identifying and resolving hardware and software issues at various levels.

• Test Automation: Experience with test automation frameworks and methodologies for developing and maintaining automated regression tests and scripts.

• Analytical and Problem-Solving Skills: Excellent analytical and problem-solving abilities to dissect complex systems, identify issues, and propose innovative solutions.

• Communication and Collaboration: Strong communication and interpersonal skills for effective collaboration with cross-functional teams and stakeholders.

• Attention to Detail: Meticulous attention to detail to catch discrepancies and ensure thorough validation of systems and processes. 

Education and experience

Typically, a bachelor's or master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field is required. 2+ years of experience in SoC/ASIC validation and debug, particularly within the context of high-performance computing and parallel programming, is highly beneficial.