Are you passionate about complex problem-solving, high-tech systems, and software engineering? We’re always happy to connect with talented physics software developers, computer scientists, simulation engineers, and those with expertise in nuclear fusion, engineering mathematics, high-performance computing, or physics. If you’re excited about shaping the future of science through innovative software, we’d love to hear from you.

Whether you’re interested in a full-time position, internship, or traineeship, feel free to reach out to us with open application to careers@ignitioncomputing.com including the following:

• CV
• Brief motivation
• Code samples or links to repositories that you have worked on, if available

About Us

At Ignition computing we specialize in high-performance computing and research software. Be it through building infrastructure for nuclear fusion research, reducing computation times through simulation acceleration or by building a platform for greenhouse optimization, we work with our clients to make the world a better and greener place.

As a developer at Ignition Computing you will provide programming and consulting services for research support and development. You will get the chance to be involved at every step of the way in the development of software.

We mainly code in Python, C++, and Fortran. Our team is all about sharing know-how and helping each other thrive. If you want to grow your skills as a developer, you’re the kind of person we’re after.

Ignition Computing is based in Eindhoven, in the Apparatenfabriek in Strijp-S. Our culture is focussed on cooperation, growth and a healthy work-life balance. At Ignition Computing we believe in an informal environment where everyone is equal. We work closely together with Afterservice, sharing not only our office, lunch, and social events but also parts of our software stack and some ongoing projects.

Projects we are currently involved in

• We play a significant role in developing the IMAS ecosystem. This is the “universal data language” for fusion research at ITER and other facilities. Recent projects include:
  • Pulse Design Simulator, an integrated modeling tool to validate ITER pulse schedules against machine limits before experiments
  • In-Pulse Data Processing at ITER. For example, visualization in the control room of the live equilibrium reconstruction from measured magnetics data
  • A Waveform-Editor for preparing simulations and experiments
  • NetCDF/Zarr cloud storage for IMAS-compatible data, first developed with our IMAS-Python library
  • SimDB, a tool to track, manage, and query simulation data based on their metadata
  • IMAS-Validator, a PyTest like tool to validate IMAS IDSs (Interface Data Structures) through the use of general and case-specific validation rules
  • IMAS-ParaView, a tool to convert complex gridded structures to the VTK format and back. Complemented with ParaView plugins that can visualize these grids, as well as other IDS data
• We develop a simulation acceleration library, PreconNet. This project uses machine learning to accelerate complex, computationally-heavy physics simulations by generating preconditioners, optimizing solver configurations and providing initial guesses for iterative solvers
• We work with clients in industry to optimize and accelerate simulation codes, build visualization tools, and solve modeling challenges

At Ignition Computing, we play a significant role in developing the IMAS ecosystem, the universal data language for fusion research at facilities like ITER. A major challenge in operating such a complex machine is processing and visualizing vast amounts of data in real-time (in-pulse) to monitor and control the plasma. This is at the core of in-pulse data processing, where data pipelines must be robust, fast, and reliable for control room operations.

This internship focuses on developing and demonstrating these live data pipelines. A pipeline could for example use magnetics measurements for equilibrium reconstruction, and be visualized in real-time using IMAS-MUSCLE3. The aim is to demonstrate how real-time processed data can be displayed and used during ITER operations, and visualized in the control room.

Ignition Computing is based in Strijp-S, Eindhoven. Our culture is focused on cooperation, growth and a healthy work-life balance. At Ignition Computing we believe in an informal environment where everyone is equal.

Responsibilities

• Conduct a study on in-pulse data processing in fusion (e.g., at WEST) and real-time visualization techniques
• Develop an example processing pipeline integrated with MUSCLE3 and Apache Kafka
• Implement a live visualization actor to display the processed data, using the Python Panel framework
• Analyze the performance, latency, and robustness of the developed pipeline for control room applications

What we ask of you

• Masters student with a background in Scientific Computing, Computer Science, or Physics
• An interest in data visualization, real-time systems, or fusion energy
• Programming skills and knowledge/interest in programming languages such as Python, C++, Fortran or MATLAB
• Fluent in written and spoken Dutch and/or English
• Willingness to learn

What you get from us

• Good support from enthusiastic team members
• The ability to work on ITER/WEST
• Free all-you-can-eat lunch, drinks and snacks
• Lots of opportunities to grow

To apply, send a brief summary of your background along with a short description of who you are to careers@ignitioncomputing.com.

Achieving commercial fusion energy depends significantly on advanced simulation tools. Startups like Pacific Fusion need trustworthy, high-fidelity codes to design fusion targets and accurately predict experimental results. FLASH, a publicly available multi-physics MHD code, has a history in astrophysics and high-energy-density physics research. Pacific Fusion launched a project to validate it for their specific Inertial Confinement Fusion (ICF) applications. Ignition Computing contributes to this goal by leveraging our computational expertise.

Our focus is on improving FLASH’s core functionality. We are tackling several key areas to ensure more stable and efficient simulations:

• Solver Reliability: We found and fixed critical bugs in the setup of linear systems, crucial for solving diffusion in MHD processes.
• AMR Accuracy: We addressed inaccuracies in how the Adaptive Mesh Refinement (AMR) system managed boundaries between different refinement levels, leading to more precise and stable outcomes.
• Linear Solver Optimization: We have boosted the performance of HYPRE’s linear solvers and their preconditioners (like Algebraic Multi-Grid methods). As a result, the solvers are more robust and converge faster.
• Equation of State (EOS) Speed: We optimized the EOS calculations and table lookups, significantly reducing a major computational bottleneck and reducing overall simulation times.

These core improvements to FLASH’s stability and speed are crucial for the progress of the wider validation project. A faster, more reliable FLASH enables the team to effectively run the necessary suite of physics benchmarks.

The validation phase compares the enhanced FLASH code against experimental results, theoretical models, and established ICF codes such as HYDRA and LASNEX. Across six challenging test scenarios, FLASH accurately models complex physics like hydrodynamic instabilities and integrated MagLIF performance. The results show strong alignment with experiments conducted on the Z-machine and with simulations scaled up to the 60 MA range, which is directly relevant to Pacific Fusion’s future goals.

By engaging Ignition Computing’s expertise in fusion simulation, we were able to significantly speed up the completion of this extensive validation effort. Few groups have the right knowledge to immediately contribute to a project like this, and Ignition was the type of low-friction, high-delivery partner we needed to make rapid progress” ~ Leland Ellison, Simulation and Modeling Lead at Pacific Fusion.

This ongoing collaboration showcases Ignition Computing’s focus: using our expertise in computational science to enhance simulation tools for demanding research areas such as fusion energy, making them more powerful, reliable, and efficient. We are happy to see that our collaboration with Pacific Fusion has helped FLASH’s successful validation, and we are looking forward to continue!

Read more at Pacific Fusion

The opinions expressed in this article are those of Ignition Computing only and do not represent Pacific Fusion’s official position.

IMAS, or Integrated Modelling & Analysis Suite, is a collection of software used at ITER that aims to provide a standardized way to represent and exchange data between different simulation codes and experimental setups. With the IMAS-ParaView tools we developed, IMAS users can load their data in ParaView, which is an open-source application for visualizing and analyzing large scientific datasets. ParaView provides powerful visualization and analysis capabilities, which can now be utilized by the IMAS community.

The release of the tool was accompanied by a two-day, on-site training at ITER that combined live demonstrations with hands-on practical sessions. We were happy with the attendance and the positive feedback, and hope the training supports participants in integrating IMAS-ParaView into their data visualization workflows.

IMAS-ParaView is open-source and can be downloaded and installed from PyPI. You can read the documentation on ReadTheDocs and view its source code on GitHub. The training material is also made available.

If you would like to hear more about this project feel free to get in touch!

The opinions expressed in this article are those of Ignition Computing only and do not represent the ITER Organization’s official position.

The international organisation ITER is building the largest tokamak in the world in France. This ‘big science’ project aims to prove useful power generation with nuclear fusion. This multi-billion USD project, built by partners in 35 nations, is one of the most complex machines ever built. Predictive modelling of this massive machine requires the coupling of many simulation tools and is a major challenge.

IMAS-Python is a pure Python library for working with IDS data structures from the IMAS Data Dictionary. This allows for a standardized data exchange between different programs used for simulations and analysis of experimental data. We have previously written about IMASPy on this site. Since then, we have worked with ITER Organization to add many features to improve the usability, performance and utility of IMAS-Python.

This new release marks a transformative moment, where IMAS-Python has surpassed the original Python IMAS interface in terms of functionality and performance and is ready to be the main Python IMAS interface. It is also one of the first ITER projects that has been released under the open-source LGPL-3.0 license on GitHub. This makes the software more accessible and easier to install. We are very happy with this decision by the ITER Organization: making the tools required for research and collaboration openly available will accelarate reaching the goal of useful power generation with nuclear fusion!

IMAS-Python can be downloaded and installed from PyPI. You can read the documentation on ReadTheDocs and view its source code on GitHub.

Although IMAS-Python is now open sourced, the core Access Layer component (imas_core) is not (yet) publicly available. This means that some functionality is disabled, but sharing data can still be done through the recently added netCDF file interface when the netCDF4 Python package is installed.

If you would like to hear more about this project feel free to get in touch!

The opinions expressed in this article are those of Ignition Computing only and do not represent the ITER Organization’s official position.

The international organisation ITER is building the largest tokamak in the world in France. This ‘big science’ project aims to prove useful power generation with nuclear fusion. This multi-billion USD project, built by partners in 35 nations, is one of the most complex machines ever built. Predictive modelling of this massive machine requires the coupling of many simulation tools and is a major challenge.

Most integrated modelling workflow tools use a directed acyclic graph (DAG) to organise operations. For co-simulation purposes, with an unknown number of iterations and many timestep loops, this paradigm is problematic. It is then better to work with semi-autonomous agents which exchange messages with each other, such as implemented in MUSCLE3, the Multiscale Coupling Library and Environment.

We have developed tooling to facilitate distributed simulations with codes in multiple languages, with distributed checkpointing, efficient communication using ITER IMAS IDSes, and support for MATLAB Simulink.

In the coming year we will build on this tooling to develop a Pulse Design Simulator, which will be used to simulate the working of the ITER tokamak. The developed infrastructure will host design simulators of various complexity and speed, tailored to the different use cases for experimental design. It will be used to verify proposed experiments against machine limits, to design better experiments and to inspire further, more detailed, simulation runs.

If you would like to hear more about this project feel free to get in touch!

The opinions expressed in this article are those of Ignition Computing only and do not represent the ITER Organization’s official position.

The SME Innovation Stimulus for Regional and Top Sectors (MIT in Dutch) encourages Small and Medium-sized Enterprises (SMEs) to develop creative initiatives. These top ten sectors in the Netherlands have been designated as those that have the potential to solve global problems and help boost the country’s economy and competitiveness. The MIT R&D AI is the dedicated stimulus program for innovative collaboration projects by SMEs to deploy innovative AI tools in the top sectors.

Our project proposal was graded on objective criteria on its potential, innovativeness, sustainability and feasibility. It received excellent marks (94.9/100 points) from the RVO. This grant will allow us to develop simulation acceleration tools for complex multiphysics problems, and deploy them in the glass industry.

Our technology- partner CelSian has decades of experience in CFD tools used for the quality prediction of glass making. We will be integrating and adapting Preconnet with CelSian’s market-leading GTM-X tool (see article picture). Preconnet is a machine-learning based tool to accelerate simulations by generating preconditioners, initial guesses and proposing solver configurations. Preconnet has been developed for plasma simulations in a previous project with Plasimo. Thanks to its versatility, Preconnet’s functionality can be extended to new simulation challenges.

The glass-melting process can take up to multiple days. This makes glass production difficult. Temperature and material fluctuations have implications for the final product. Product quality issues, such as bubbles in glass, may arise. This can lead to the loss of large batches of glass, totaling multiple days of production. New developments in machine learning and simulation provide more accurate predictions of glass quality.

The project is expected to result in improved quality prediction models leading to substantial economic and environmental gains. Through more efficient feed-stock use and energy consumption, each glass oven could save energy costs in the range of € 250.000 to € 1.250.000 per year as well as reducing CO2 and nitrous oxide emissions. This undeniably positive expected impact fits with our vision of supporting the sustainable industry of the future.

The global fusion landscape is undergoing a profound change. The ambition of harnessing “the energy of the Sun and stars” was traditionally pursued through public projects which were government or university-funded, culminating in megaprojects such as ITER. In recent years, new and daring private initiatives have managed to gather substantial private investment, to seek new approaches to realizing the fusion dream. In order to encourage synergy between public and private initiatives, the ITER organization hosted a three-day workshop.

Some startups are working on the well-established magnetic confinement fusion pathways such as Stellarators and Tokamaks while harnessing more agile or more integrated organizational approaches and technological advances in superconductor technologies. Other startups are exploring more exotic avenues, such as inertial confinement fusion. Regardless of the approach, the first step is to test the ideas in simulation. Daan van Vugt spoke to Computer Weekly about five key areas in fusion research and development where high-performance and scientific computing make an impact.

Ignition computing celebrates its new membership of the FuseNet Association! This association stimulates Fusion education in Europe. Fusenet helps students and educators alike through education development and mobility programs and it forges bonds between universities, research institutes and industry.

As many at Ignition Computing have themselves profited directly from Fusenet’s help, they look forward to providing a place for students to learn about software development for Nuclear Fusion.

Taxila is a comprehensive platform that serves as a hub for trainers and trainees seeking to explore a wide range of training opportunities, learning resources, and teaching materials. Whether you are interested in Research Data Management, Research Software Management, Open Science, or other related fields like High Performance Computing, Taxila provides a centralized space where you can discover upcoming workshops, seminars, conferences, and webinars.

Taxila continues to grow in multiple aspects. The code base is based on ELIXIR’s TeSS and is in active development to ensure a seamless user experience, incorporating user feedback and introducing new features to improve the platform’s functionality. On the other hand, Taxila’s reach is constantly expanding as an increasing number of training providers join the platform. This growing network of providers enriches the diversity and quality of the training materials and opportunities available on Taxila, creating a dynamic ecosystem where trainers and trainees can connect, collaborate, and share knowledge.

More info can be found on the Taxila site.