Research Software Engineer - Scientific Computing & GPU Infrastructure

Marvel Fusion Inc.; Munich (ND); On-site; USD 78,000 - 123,000; Full time

Benefits offered by this job

Public Transport Subsidy

Employee Incentive Program

Phone Bill Reimbursement

Health Insurance

Professional Learning & Development

Relocation Support

Company Pension Plan

30 Days of Vacation

Modern Office in Munich

Healthy office snacks

Team Events

Job summary

Marvel Fusion Inc. in Munich is seeking a passionate software engineer to advance scientific Python software for fusion research. You will shape computational approaches, optimize numerical performance, and enable GPU-accelerated workflows on NVIDIA and cloud platforms.

Join a multidisciplinary team of scientists and engineers tackling complex physics and engineering problems. Expected to contribute to testing, CI, documentation, and robust data handling while communicating clearly in English.

Qualifications

• Degree in Computer Science, Physics, Applied Mathematics, Engineering or related field.
• Strong Python software engineering skills with maintainable code in an existing codebase.
• Experience with scientific/numerical computing and verifying numerical accuracy.
• Experience profiling workloads and GPU computing with CuPy, PyTorch, JAX or CUDA.
• Experience deploying Python/GPU workloads on remote Linux servers or cloud infra with reproducible environments.
• Structured approach to debugging, version control, automated testing and documentation.
• Self-driven, collaborative, with excellent English written and verbal communication.

Responsibilities

• Develop and maintain scientific Python software translating research requirements into features.
• Work with scientists to define acceptance criteria and verify results against references.
• Profile workloads and compare algorithms across CPU/GPU implementations with benchmarks.
• Prepare and deploy workloads on NVIDIA GPU servers or AWS using reproducible environments.
• Support multiple users with access controls, job scheduling, and data handling.
• Connect computational services to internal apps via stable APIs for job submission and results.
• Improve testing, CI, packaging, and documentation; deliver assigned tasks.

Skills

Python

Numerical computing

Profiling

GPU computing

Testing & CI

English communication

Debugging

Collaboration

Education

CS/Physics/Math/Engineering degree

Tools

CuPy

PyTorch

JAX

CUDA

AWS

NVIDIA GPU servers

Job description

Who we are

Founded in 2019, Marvel Fusion is Europe’s leading fusion energy company, uniting 120 scientists, engineers, and entrepreneurs across our locations in Munich and Colorado.

Backed by over €385 million in public and private funding, we’re driven by a shared mission: to deliver clean, abundant energy to the world.

Why Marvel Fusion

By joining us, you will be:

• Solving one of the most complex technological challenges known to humanity, harnessing Fusion on Earth
• Part of a highly purpose-driven team working on providing the world with clean, safe and abundant energy
• Working alongside world-leading scientists and entrepreneurs in the field of Fusion
• Part of a start-up where growth on a company and individual level is the default

The impact you will make

As part of our multidisciplinary research and engineering team, you will shape how we tackle computational challenges in fusion research. You will combine your software expertise with the team’s knowledge of physics and engineering to develop new approaches, make technical decisions and turn ideas into reliable scientific software. From advancing numerical performance to enabling shared GPU computing on NVIDIA servers or AWS, you will own meaningful technical challenges and contribute directly to our scientific progress.

Your responsibilities

• Develop and maintain scientific Python software, turning research requirements into reusable, tested features and clear interfaces.
• Work with scientists to define acceptance criteria and verify numerical results against reference cases and agreed accuracy requirements.
• Use profiling to identify bottlenecks; compare algorithms, code translations and CPU/GPU implementations through reproducible benchmarks.
• Prepare and deploy scientific workloads on NVIDIA GPU servers or AWS using reproducible environments and automated execution workflows.
• Work with IT to support multiple users through appropriate access controls, job scheduling, resource allocation, monitoring and result storage.
• Connect computational services to internal applications through stable APIs for job submission, status tracking and result retrieval.
• Improve testing, continuous integration, packaging and documentation, and independently carry agreed development tasks through to delivery.

What you bring

• A degree in Computer Science, Physics, Applied Mathematics, Engineering or a related field, or equivalent practical experience.
• Strong Python software engineering skills and a track record of independently delivering maintainable software within an existing codebase.
• Experience with scientific or numerical computing, including testing numerical accuracy and collaborating with domain experts.
• Practical experience profiling computational workloads and using GPU computing frameworks such as CuPy, PyTorch, JAX or CUDA.
• Hands-on experience deploying scientific Python or GPU workloads on remote Linux servers or cloud infrastructure, including reproducible environments, job execution and data handling.
• A structured approach to debugging, version control, automated testing and documentation, with the ability to make pragmatic engineering decisions.
• A self-driven, collaborative working style and excellent written and verbal communication skills in English.

Nice to have

• Direct experience with AWS GPU services or shared NVIDIA GPU infrastructure, including containers, workload scheduling or monitoring.
• Expertise in CUDA, multi-GPU computing, high-performance computing or performance-portable numerical software.
• Experience integrating scientific applications with compute services, or a background in optics, photonics or experimental data analysis.
• Experience evaluating AI-based acceleration of scientific workloads against accuracy and runtime benchmarks.

What we offer

Our mission is motivation enough for many. But we also support our team with benefits that foster wellbeing and growth:

• Public Transport Subsidy
  
• Employee Incentive Program
• Phone Bill Reimbursement
• Wellpass or Complimentary Private Health Insurance
• Professional Learning & Development Budget
• Relocation Support
• Company Pension Plan
• 30 Days of Vacation plus additional half day on 24. & 31.Dec.
• Modern Office in the center of Munich
• Healthy office snacks
• Regular Team Events

For more information on how we process your application data, please review our privacy notice.