Skip to content
View JohannesBroens's full-sized avatar
🏈
🏈

Highlights

  • Pro

Block or report JohannesBroens

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JohannesBroens/README.md

Johannes Brøns Christensen

Professional Profile

Machine Learning and Data Scientist with a Bachelor of Science from the University of Copenhagen (DIKU). Experienced in the development of data pipelines, analytical models, and visualization frameworks within industrial and regulated environments. Technical expertise includes deep learning, high-performance computing, and statistical modeling, with a focus on transforming complex datasets into actionable operational insights.


Technical Skills

  • Programming Languages: Python (PyTorch, NumPy, Pandas, TensorFlow), C, SQL, R
  • Machine Learning & Analytics: PyTorch Lightning, Scikit-Learn, XGBoost, MLflow, Time-series analysis, Anomaly detection
  • Data Engineering & Visualization: ETL pipelines, Alteryx, Power BI, Matplotlib, Plotly, Grafana, SquaredUp
  • Systems & DevOps: Linux/UNIX, Git, Docker, CUDA, OpenMP, MLOps principles, REST APIs

Professional Experience

Novo Nordisk | Data & Analytics Specialist / Associate IT Operation Manager

January 2024 – January 2026

  • Developed Python-based analytics tools for operational insights in IT/OT environments.
  • Engineered SQL data models and automated data preparation workflows for large-scale industrial reporting.
  • Conducted statistical analysis on incident and root-cause trends to drive infrastructure reliability.
  • Implemented monitoring solutions integrating SCOM, SquaredUp, and Grafana for critical infrastructure.

Selected Projects

Hidden Markov Model for Visual Attention Analysis

Developed a comprehensive Hidden Markov Model (HMM) to simulate and analyze visual attention patterns using neural spike data.

  • Implemented forward simulation algorithms to generate synthetic neural activity.
  • Designed exact inference via variable elimination and message passing algorithms.
  • Applied approximate inference using logistic regression and implemented a hard-assignment EM algorithm for parameter learning.
  • Validated the framework across both simulated and real-world neural datasets.

Satellite Image Segmentation for Arctic Mapping

Collaborative research project with NASA and the University of Copenhagen (DIKU and IGN) focused on Greenland’s coastline.

  • Developed a U-Net architecture for high-resolution segmentation of satellite imagery.
  • Achieved a spatial accuracy of 0.87 IoU (Intersection over Union).
  • Optimized inference speed for large-scale geospatial datasets.

Automated Quality Control of Meteorological Data (BSc Thesis)

Conducted in collaboration with the Danish Meteorological Institute (DMI) to automate the validation of sensor data.

  • Designed and benchmarked three distinct machine learning architectures to detect faulty sensor readings.
  • Documented and analyzed DMI's ETL pipelines and manual validation processes.
  • Developed a solution to reduce manual overhead for climatologists through automated anomaly detection.

Production Line Anomaly Detection

Developed predictive maintenance tools for manufacturing environments.

  • Built an LSTM-based (Long Short-Term Memory) recurrent neural network for early equipment failure detection.
  • Integrated models with OPC-UA data streams for real-time signal monitoring and processing.

High-Performance ML Implementation (ML-in-C)

Personal project focused on the fundamentals of computational efficiency in neural networks.

  • Implemented a Multi-Layer Perceptron (MLP) from scratch in C.
  • Utilized OpenMP for CPU multi-threading and CUDA for GPU acceleration to optimize training performance.

Education

B.Sc. in Machine Learning & Data Science

University of Copenhagen (2019 – 2023)

  • Key Coursework: Algorithms & Data Structures, High Performance Programming, Probability Theory & Statistics, Linear Algebra, Advanced Deep Learning, Models for Complex Systems.
  • Specialization: Medical Image Analysis, Lebesgue Integral & Measure Theory, and Satellite Segmentation.

Academic Service & Teaching

  • Teaching Assistant: Empirical Methodologies & Theory of Science (2021 – 2023). Guided students through research methodology and academic reasoning.
  • Teaching Assistant: Programming Intro Course (2021). Supported first-year students in Python and programming fundamentals.
  • Student Mediator: Represented the Machine Learning & Data Science program at the Department of Computer Science (DIKU), coordinating outreach and curriculum communication.

Contact and Links

Pinned Loading

  1. eu-electricity-forecasting eu-electricity-forecasting Public

    XGBoost-based day-ahead price forecasting pipeline for European electricity bidding zones (DK1, DK2, NO2, SE3, SE4, DE-LU).

    Jupyter Notebook

  2. ml-language-playground ml-language-playground Public

    A multi-language machine learning benchmark comparing neural network implementations across C, Rust, and Python. Two model families --- MLP and CNN (LeNet-5) --- are each implemented identically in…

    Rust 1

  3. x11-mcp x11-mcp Public

    Native X11 MCP server for Claude Code — bare-metal desktop automation with python-xlib, python-uinput, and AT-SPI

    Python

  4. x11-mcp-voice x11-mcp-voice Public

    Python

  5. Machine-Learning-based-Automatic-Quality-Control-of-Greenlantic-Climate-Data Machine-Learning-based-Automatic-Quality-Control-of-Greenlantic-Climate-Data Public

    Small snippets of my bachelors project I made in coorporation with The Danish Meteorological Institute

    Jupyter Notebook 1

  6. Automatic-Satelitte-Island-Discovery Automatic-Satelitte-Island-Discovery Public

    Project outside of course scope at (BSc) Machine Learning and Data Science education programme. Colab between NGI and DIKU at University of Copenhagen.

    Jupyter Notebook 1