The Department of Physics at Carnegie Mellon University (CMU) invites applications for a Research Software Engineer (RSE) position, involving the development, testing and deployment of scalable analysis and machine learning (ML) pipelines to enable science with the Compact Muon Solenoid (CMS) experiment.

The CMS experiment at the Large Hadron Collider (LHC) in Geneva, Switzerland, is a multi-purpose high energy physics (HEP) detector built and operated by a collaboration of 3000 scientists from all over the world. It collects data generated in the collisions delivered by the accelerator, with a complex data mining pipeline that involves the utilization of ML algorithms as well as cutting-edge data analysis techniques. So far the experiment has collected hundreds of petabytes of data, and it is expected to reach exabytes in the next decade.

The RSE will be integrated in the research activities of the local CMU CMS group, as well as be in the larger CMS and HEP communities. At CMU, the RSE will work on establishing best software engineering practices, leveraging cross-disciplinary computational techniques through the collaboration with scientists from other domains, as well as with core computing experts. The RSE will work closely with faculty, student/postdoctoral researchers, and technical staff to provide computational expertise in algorithm and software design, in order to produce high-quality and sustainable research code and to train the HEP community in more modern computing tools and more professional development approaches. The RSE will have the opportunity to create partnerships with members of the CMU Software Engineering Institute, the Machine Learning Department, the Department of Statistics & Data Science, and the Department of Electrical and Computer Engineering, which all rank among the best in the world in their respective fields. The RSE will also have the chance to engage with the NSF AI Planning Institute for Data-Driven Discovery and of the LINCC Frameworks program (). More specifically, responsibilities will include (but are not restricted to):

• Development, maintenance, and management of open-source projects to facilitate columnar analysis with HEP data, including support of the user community

• Study and application of complex ML algorithms to the solution of physics challenges in CMS and HEP

• Establishment of proper practices to build pipelines for ML training using the most popular ML frameworks like Tensorflow and PyTorch

• Enabling the acceleration of ML inference with co-processors like GPUs and FPGAs, to allow for the usage of ML in real-time analysis

A strong background in scientific programming, academic research, and an interest in HEP are required. Essential qualifications are:

• A PhD in physics or in an another quantitative discipline

• A minimum of 4 years as RSE or equivalent, including industry experience

• Strong software engineering skills with demonstrated successes working in a collaborative environment, as well as independently

• Curiosity to learn new concepts and technologies beyond the area of core domain knowledge

• Ability to lead software projects and manage heterogeneous teams

Preferred qualifications are:

• GPU programming experience and/or experience with FPGAs

• Experience working in an academic research environment

• Expertise in academic research

• Background in HEP

Applications, including a cover letter, curriculum vitae, publication list, and a statement that describes your past experience and future interests should be submitted via this Interfolio site: . In addition, candidates should arrange for three letters of recommendation to also be uploaded to the Interfolio website. For questions about the position, please contact Matteo Cremonesi at mcremone@andrew.cmu.edu.

The position is available as soon as possible but remains open until filled. Applications received by May 31, 2023 will receive full consideration.