HPC Development Support Engineer - Customer Residency

## 🎯 Role Overview

Highly skilled HPC professional specialising in NVIDIA GPU development practices, embedded within the customer's development teams. This resident role focuses on supporting application migration and optimization on new HPC infrastructure, acting as a trusted technical advisor and bridge between customer teams and infrastructure partners.

## 📊 Context

**Deployment Model**

On-site residency Pau within customer's development teams

**Infrastructure Focus**

New HPC environment with NVIDIA GPU acceleration

**Partnership Ecosystem**

Collaboration with Dell infrastructure and NVIDIA compiler support teams

**Mission**

Enable successful application migration, optimization, and long-term operational excellence

## 🔧 Key Responsibilities

### Application Migration and Optimization

• Assist in migrating applications to the new HPC environment, ensuring compatibility and performance
• Profile and fine-tune HPC applications for the production system
• Identify performance bottlenecks and optimization opportunities
• Ensure applications leverage GPU acceleration effectively

### Technical Expertise

Troubleshoot and resolve issues related to

• Support continuous improvement initiatives for application lifecycle management
• Provide hands-on technical guidance for complex HPC challenges

### Trusted Advisor Role

• Provide strategic guidance on development and optimization best practices for HPC environments
• Recommend tools, frameworks, and processes to improve scalability and reliability
• Share industry best practices and emerging technologies
• Mentor customer teams on HPC development methodologies

### Collaboration and Governance

• Participate in regular meetings with the customer to review progress and align priorities
• Act as a liaison between customer development teams and Dell infrastructure technical experts
• Act as a liaison between customer development teams and NVIDIA compiler support
• Facilitate knowledge transfer and cross-team communication
• Document technical decisions, recommendations, and lessons learned

### DevOps Enablement

• Assist in designing and implementing CI/CD pipelines tailored for HPC applications
• Integrate automated testing and deployment workflows
• Support DevOps best practices adapted to HPC constraints and requirements
• Enable reproducible builds and deployment automation

## 💻 Required Technical Skills

### HPC & GPU Development

**NVIDIA GPU Programming**

CUDA, profiling tools (Nsight, nvprof)

**HPC Application Optimization**

Performance tuning, parallelization strategies

**Profiling & Benchmarking**

Application performance analysis and optimization

**Compiler Expertise**

NVIDIA HPC SDK, GPU-accelerated libraries (cuBLAS, cuFFT, NCCL)

### Software Development

**Programming Languages**

C/C++, Python, Fortran (asset)

**Parallel Programming**

MPI, OpenMP, GPU acceleration

**Version Control**

Git, GitLab/GitHub workflows

**Build Systems**

CMake, Make, containerization (Docker, Singularity)

### HPC Infrastructure

**Linux Systems**

Deep knowledge of HPC Linux environments

**Job Schedulers**

Slurm, PBS, LSF experience

**Storage Systems**

Parallel file systems (Lustre, GPFS)

**Networking**

High-speed interconnects (InfiniBand, RoCE)

### DevOps & Automation

**CI/CD Tools**

Jenkins, GitLab CI, GitHub Actions

**Container Technologies**

Docker, Singularity/Apptainer for HPC

**Scripting**

Bash, Python for automation

**Infrastructure as Code**

Ansible, Terraform (asset)

## ✅ Required Profile

### Essential Experience

• ✅ **5+ years in HPC environments** (supercomputing centers, research labs, or industry HPC)
• ✅ **Strong NVIDIA GPU development experience** (CUDA programming, optimization)
• ✅ **Application porting and optimization** on HPC systems
• ✅ **Profiling and performance analysis** expertise
• ✅ **Linux system administration** in HPC contexts

### Technical Competencies

• ✅ Deep understanding of **GPU architecture and memory hierarchy**
• ✅ Experience with **parallel programming models** (MPI, OpenMP, CUDA)
• ✅ Knowledge of **HPC software stack** (compilers, libraries, schedulers)
• ✅ Ability to **troubleshoot complex technical issues** across the stack
• ✅ Experience with **DevOps practices** in scientific computing environments

### Soft Skills & Working Style

• ✅ **Excellent communication skills** - ability to explain complex technical concepts
• ✅ **Customer-facing experience** - comfortable working embedded in client teams
• ✅ **Collaborative mindset** - bridge between multiple technical teams
• ✅ **Autonomy and initiative** - self-directed problem-solving
• ✅ **Pedagogy and mentoring** - ability to transfer knowledge effectively
• ✅ **Adaptability** - comfortable in dynamic, evolving environments

### Highly Valued Experience

• 🎯 Previous **residency or consulting roles** in customer environments
• 🎯 Experience with **NVIDIA HPC SDK** and compiler toolchain
• 🎯 **Dell/HPE HPC infrastructure** knowledge
• 🎯 **Scientific computing applications** (CFD, molecular dynamics, climate modeling, etc.)
• 🎯 **CI/CD for HPC** - experience with automated workflows for scientific applications
• 🎯 **Multi-vendor ecosystem** experience (NVIDIA, Intel, AMD)

## 📍 Working Arrangements

### Location & Work Model

• **On-site residency** at PAU
• Embedded within customer development teams
• Close collaboration with infrastructure teams (NVIDIA support)

### Language Requirements

• **Fluent English** required (technical documentation, meetings, presentations)
• **French** proficiency (asset, depending on customer location)

### Additional Requirements

• Ability to work in **collaborative, cross-functional teams**
• Comfortable with **regular customer interaction** and reporting
• Willingness to travel occasionally for training or vendor meetings

## 🎯 Success Criteria

• Successful migration of critical applications to new HPC infrastructure
• Measurable performance improvements in optimized applications
• Strong customer satisfaction and trust relationship
• Effective knowledge transfer to customer teams
• Smooth collaboration between all technical stakeholders