Data Reduction Physicist

Job Posting Organization:
CERN, the European Organization for Nuclear Research, is at the forefront of scientific research and technological innovation. Established in 1954, CERN has grown to become one of the world's largest and most respected centers for scientific research, employing over 2,500 staff members and hosting thousands of scientists from around the globe. The organization operates in multiple countries, with its headquarters located in Geneva, Switzerland. CERN's mission is to push the boundaries of knowledge and understanding of the universe through groundbreaking research in particle physics, and it fosters a collaborative environment that encourages innovation and diversity among its workforce.

Job Overview:
The position of Data Reduction Physicist is pivotal in the NGT Real-time Reconstruction Revolution project, which aims to transform the CMS High-Level Trigger into a facility capable of processing up to 750 kHz of events with offline-quality reconstruction. The successful candidate will lead the design of two complementary data tiers: a low-level reconstructed format that allows for significantly higher trigger rates than conventional raw-data outputs while maintaining the option for future reprocessing, and a compact analysis format that captures the full stream of online reconstructed events. This role requires a deep understanding of data structures and the ability to innovate in data reduction techniques, ensuring that physics performance and analysis flexibility are preserved while pushing the limits of data processing capabilities.

Duties and Responsibilities:
The Data Reduction Physicist will be responsible for designing next-generation data structures that maximize data reduction efficiency to accommodate the high event rates of 750 kHz. Key responsibilities include ensuring that all data formats are accelerator-native and optimized for high-throughput GPU processing. The physicist will also build an end-to-end framework to rigorously quantify the impact of lossy compression, establishing clear metrics, reference analyses, and automated regression tests. Additionally, benchmarking compression and decompression under realistic workloads, including CPU/GPU cost, I/O throughput, memory footprint, and latency, will be essential. The role also involves advancing lossless compression techniques, leveraging R³-reconstructed objects, and pioneering AI/ML methods to enhance data reduction processes.

Required Qualifications:
Candidates must demonstrate significant contributions to trigger and/or reconstruction in High Energy Physics (HEP) or comparable high-throughput scientific software. A practical understanding of end-to-end HEP experiment operations, from detector readout to reconstruction, calibrations, datasets, and final physics results is essential. Experience in large international collaborations, including code review, CI/CD, and documentation, is advantageous. Knowledge of LHC experiments and their data formats, as well as expertise in data compression techniques (both lossless and lossy), is highly desirable. Familiarity with AI/ML methods, particularly autoencoders, and proficiency in GPU programming and heterogeneous computing will be beneficial.

Educational Background:
The ideal candidate will possess a professional background in Physics or a related field, with either a Master's degree accompanied by 2 to 6 years of post-graduation professional experience, or a PhD with no more than 3 years of post-graduation professional experience. Candidates must also be nationals of a CERN Member or Associate Member State and should not have previously held a CERN fellow or graduate contract.

Experience:
The position requires a moderate level of experience, specifically 2 to 6 years of post-graduation professional experience for those with a Master's degree, or up to 3 years for PhD holders. Candidates should have a proven track record in relevant fields, particularly in high-throughput scientific software and HEP experiment operations.

Languages:
Proficiency in spoken and written English is mandatory, with a commitment to learn French being a valuable asset. This bilingual capability will enhance communication within the diverse international environment at CERN.

Additional Notes:
The contract duration for this position is 24 months, with the possibility of extension up to a maximum of 36 months. The working hours are set at 40 hours per week, and the position is fully onsite. The target start date for this role is January 1, 202

• The job reference number is EP-CMS-TDQ-2026-146-GRAP. Compensation includes a monthly net stipend ranging from 6,372 to 7,004 Swiss Francs, depending on the candidate's degree. Additional benefits include 30 days of paid leave per year, comprehensive health insurance coverage, family allowances, a relocation package, and opportunities for on-the-job and formal training, including language classes.