Satheshkumar K.
Background
PhD from Western Norway University of Applied Sciences. My research activities primarily revolved around machine learning and medical image analysis, with a particular focus on design methodologies in deep learning for efficient use of data. Currently, I am employed as a data scientist at Lerøy Seafood, a global leader in the seafood industry. My primary responsibilities include MLOps and LLMOps infrastructure, along with integrating machine learning models and large language models into operational workflows. These efforts aim to enhance business efficiency and support data-driven decision-making across the organization.
Experience
Data scientist
Lerøy Seafood, Norway | Present
- Built MLOps infrastructure in Databricks, automating model deployment and launching two production-grade salmon price prediction models
Partner
AkademiX | July 2023 - Present
Co-founded a consulting company focused on artificial intelligence solutions and knowledge sharing.
Assistant professor II
Western Norway University of Applied Sciences | 2021
Held a 20% position during PhD studies for teaching ML course for third year BSc students.
Data scientist
Bouvet ASA, Bergen | 2019
Worked part-time during my masters.
Researcher
University of Bergen, Department of Biomedicine | 2019-2020
Teaching assistant
Western Norway University of Applied Sciences | 2015 - 2018
Assisted in teaching various software engineering courses. Conducted lab sessions and graded assignments.
Summer Intern
Capgemini | 2018
Summer Intern
Nordea Liv | 2017
Internship
Vizrt, Bergen, Norway | 2016
Education
PhD in Computer Science
Western Norway University of Applied Sciences | 2020-2023
Thesis: Deep learning in medical image analysis: efficient use of data and radiological expertise
MSc in Software Engineering
University of Bergen | 2019
Thesis: Deep transfer learning in medical imaging
BSc in Information Technology
Western Norway University of Applied Sciences | 2017
Skills
Languages
Primary programming language
Tools
Publications
fastMONAI: a low-code deep learning library for medical image analysis
S. Kaliyugarasan, A.S Lundervold
Software Impacts, 2023
Source codeMulti-Center CNN-based spine segmentation from T2w MRI using small amounts of data
S. Kaliyugarasan, A.S Lundervold and others
20th IEEE International Symposium on Biomedical Imaging (ISBI), 2023
Source codeLAB-Net: Lidar and aerial image-based building segmentation using U-Nets
S. Kaliyugarasan, A.S Lundervold
Nordic Machine Intelligence, second place 2022 NORA MapAI competition
Source codeFully automatic whole-volume tumor segmentation in cervical cancer
E. Hodneland, S. Kaliyugarasan et al.
Cancers, 2022
Teaching
Teaching Experience
Courses
DAT158: Machine learning engineering and advanced algorithms (Fall 2021)
Supervision
MSc Projects (co-supervision)
2020-2022: A workflow-integrated brain tumor segmentation system based on fastai and MONAI
BSc Projects (co-supervision)
- 2024: Large language models and fish health
- 2025: Deep learning for quality control of fish fillets
Talks and travels
Lerøy og kunstig intelligens
Inspirasjonsdag for realfagselever, Western Norway University of Applied Sciences, Bergen, Norway
November 4, 2024
Hvordan jobber vi med kunstig intelligens?
Springbrett, Grieghallen, Bergen, Norway
September 24, 2024
Lerøy og kunstig intelligens
Lerøy motivasjonssamling, Scandic Flesland, Bergen, Norway
September 7, 2024
Kunstig intelligens i hverdagen
Tekna, VilVite, Bergen, Norway
April 29, 2024