Middle/Senior C++/Python Engineer (High Performance Computing)

We are seeking a skilled and motivated HPC Developer to join our team and contribute to the development of the computational core for digital twin modelling software. This module will perform high-performance grid calculations and advanced operations on GPUs and CPUs using state-of-the-art parallel computing techniques. You will play a critical role in designing, implementing, and optimizing asynchronous workflows with load balancing, memory allocation, and cross-platform execution capabilities.

Key Responsibilities:

• Design and implement high-performance computation modules for grid operations.
• Develop algorithms optimized for both CPU and GPU architectures using modern frameworks like CUDA, OpenCL, OpenMP, SYCL, or MPI.
• Implement task-based parallelism and asynchronous execution strategies.
• Collaborate with the orchestration team to integrate task scheduling and resource management systems.
• Profile and optimize memory allocation, load balancing, and data movement across CPU and GPU.
• Ensure efficient communication between distributed processes using MPI or similar frameworks.
• Develop portable code that supports multiple hardware configurations (NVIDIA GPUs, AMD GPUs, CPUs).
• Leverage frameworks like SYCL, Kokkos, or OpenCL for cross-platform compatibility.
• Work closely with the backend and visualization teams to integrate the computation module with the overall Odin architecture.
• Implement APIs for seamless interaction between the computation module and other components.
• Develop unit tests, benchmarks, and validation procedures for computational accuracy and performance.
• Troubleshoot and debug issues across heterogeneous systems.
• Document code, algorithms, and design decisions to ensure maintainability and knowledge sharing.

Experience and Qualification:

• 3+ years of experience in HPC, parallel computing, or numerical simulations.
• Proven experience in GPU programming (CUDA, HIP, or OpenCL).
• Hands-on experience with CPU parallelism using OpenMP, MPI, or similar frameworks.
• Familiarity with asynchronous workflows and task orchestration frameworks
• Bachelor’s or Master’s degree in Computer Science, Computational Physics, Applied Mathematics, or related fields.
• Programming Languages: Proficiency in C++/Python. Rust or Julia experience is a plus.
• Parallel Computing: Strong understanding of GPU programming (CUDA, HIP, or SYCL) and multi-threaded CPU programming (OpenMP, MPI).
• Memory Management: Expertise in optimizing memory allocation, data movement, and unified memory (e.g., CUDA UVM).
• Load Balancing: Experience with dynamic and static load-balancing techniques in HPC systems.
• Experience with orchestration tools
• Strong problem-solving skills and ability to work in a multidisciplinary team.
• Excellent communication skills for cross-team collaboration and documentation.
• Passion for innovation and a drive to push technological boundaries.

Preferred Qualifications:

• Experience in developing simulation or visualization software for oil and gas or similar industries.
• Familiarity with modern containerization technologies (e.g., Docker, Kubernetes).
• Knowledge of advanced numerical methods like finite element/volume methods, deep learning or deep neural operators.