• Part of the Applications team at Sopra Steria with focus on machine learning and cloud development.• On project with the UNICAN team at NTNU to develop state-of-the-art, easily accessible AI solutions for digital pathology.• Member of the Generative AI board in Sopra Steria's rAIse program.

• Developing chatbot for AI-assisted planning during drilling and well operations.

• In-charge of summer project responsible for three software developer interns.• Developed generative AI-accelerated prototype to assess a building's environmental sustainability.

• Part of Medical Image Analysis research group.• Consulted on numerous research projects and grant applications, either through tutoring colleagues, implementing components in algorithm or deployment design, statistical analysis in assessment of trained models, or development of accessible technologies.• Key contributor to the FastPathology open software project in C++ using Qt5 and FAST.• DevOps responsible for open-source clinical software, Raidionics, enabling automatic segmentation of pre- and postoperative brain tumors and generation of standardized clinical report.• Developed open software plugin enabling cloud-based deployment of AI-solutions for digital pathology, FP-DSA-plugin.• Developed 4 applications demonstrating AI-based medical 3D image segmentation, using Gradio and hosted on Hugging Face Spaces.• Developed open python package GradientAccumulator to enable gradient accumulation in TensorFlow 2 released on PyPI.• Codeveloped a python package to enable rapid stain normalization for histopathological images, torchstain, supporting PyTorch, TF, and NumPy.

• Part of Medical Technology research group.• Lead SINTEF-funded project to enable code-free development and deployment of deep segmentation models for computational pathology:- Trained clinical pathologists with no background in programming or deep learning to train and deploy his own convolutional neural network for semantic segmentation of gigapixel histopathological images.- Orchestrated collaboration with researcher Ilya Belevich and lead developed of MIB at University of Helsinki. Final work published here: https://www.frontiersin.org/articles/10.3389/fmed.2021.816281/fullUseful summary of work and code repository here: https://github.com/andreped/NoCodeSeg• Co-supervised 1 Master's student working on semantic segmentation of lung lobes in CTs using memory-efficient 3D U-Net architecture capable of using entire CT volume as input, enabled through seperable convolutions mixed precision in PyTorch-Lightning.• Contributed to several funding applications on various topics with focus on software as a medical device and use of AI for medical applications. Contributed strongly to the AI, software, and statistics work packages, of which multiple achieved funding from the Norwegian Research Council.• Performed statistical analysis and aided in method development and consulted in research activities, mainly focused on machine learning and computer aided designs, such as:- Supervised segmentation of brain tumors in MRIs - 5 separate papers (ex: https://www.frontiersin.org/articles/10.3389/fradi.2021.711514/full).- Supervised lymph node segmentation in CTs (https://arxiv.org/abs/2102.06515).- Unsupervised detection of adverse events from free-text (https://github.com/andreped/adverse-events).- Responsible for statistical analysis for nanobubble-guided cancer treatment study (https://doi.org/10.1016/j.ultrasmedbio.2020.12.026).

• Research and deployment of methods in computational pathology to produce a software for assisting pathologists in clinical practice.• Paper 1: FastPathology: An Open-Source Platform for Deep Learning-Based Research and Decision Support in Digital Pathology (2021). IEEE Access. https://ieeexplore.ieee.org/document/9399433• Paper 2: H2G-Net: A multi-resolution refinement approach for segmentation of breast cancer region in gigapixel histopathological images (2022). Frontiers in Medicine. https://www.frontiersin.org/articles/10.3389/fmed.2022.971873/full• Supporting articles:- Code-Free Development and Deployment of Deep Segmentation Models for Digital Pathology (2022). Frontiers in Medicine. https://www.frontiersin.org/articles/10.3389/fmed.2021.816281/full- High Performance Neural Network Inference, Streaming, and Visualization of Medical Images Using FAST (2019). IEEE Access. https://ieeexplore.ieee.org/document/8844665/

• In charge of programming workshop in Python & MATLAB, running each fall.• Tailored for students participating in the courses: FYS-1001 Mechanics & FYS-2006 Signal Processing.

• Implemented end-to-end pipeline for lung nodule cancer screening in CTs, from rawimage format to inference & visualization in software prototype.• Implemented simple morphological-based algorithm for automatic lung segmentation in CTs.• Implemented semi-supervised seed-based algorithms for tumor segmentation in CTs.

• Teached kids 6-15 years old in topics like mathematics, natural science, & gymnastics.