PeopleofSuperintelligence

Researcher

Harmen Potgieter

Also known as Hermanus L. Potgieter

Harmen Potgieter is an AI researcher whose work studies generalization in deep neural networks and computer vision for agriculture. He completed a Master of Computer and Electronic Engineering at North West University in December 2024, researching batch normalization in convolutional neural networks while affiliated as an MEng student with the Centre for Artificial Intelligence Research (CAIR). In 2024, he coauthored papers on batch-normalization representations for SACAIR and on network fragmentation as a complexity measure for the SciForDL workshop at NeurIPS. In January 2025, he began a PhD in Computer and Electronic Engineering at North West University and a software developer role at MUST Deep Learning. As of October 2026, he is a PhD student at North-West University based at the MUST Research Office in Hermanus, studying real-time cattle identification with deep vision models.

Now at
MUST Deep Learning
Badges

Timeline

  1. 2023–2024

    An MEng at North West University

    From January 2023, he studies for a Master of Computer and Electronic Engineering at North West University and completes the degree in December 2024. His master's research examines the effect of batch normalization on the generalization ability of convolutional neural networks.

  2. 2024

    MEng student at CAIR

    In 2024, he is an MEng student at the Centre for Artificial Intelligence Research (CAIR), where he publishes 2024 work on image-classification models and predicting generalization in deep learning.

  3. 2024

    Batch normalization in CNNs

    In 2024, he coauthors “Impact of Batch Normalization on Convolutional Network Representations” with Coenraad Mouton and Marelie Davel for the Southern African Conference for Artificial Intelligence Research (SACAIR). The study contrasts image-classification models trained with and without BatchNorm and finds that models with BatchNorm tend to show more advantageous clustering in hidden representations.

  4. 2024

    Testing network fragmentation

    He coauthors the 2024 paper “Is network fragmentation a useful complexity measure?” with Coenraad Mouton and four colleagues for the Scientific Methods for Understanding Deep Learning (SciForDL) workshop at NeurIPS, posted on arXiv in November 2024. The study uses a fragmentation-based complexity measure and reports good performance on the PGDL generalization benchmark.

  5. Since 2025

    PhD study and software development

    In January 2025, he starts a PhD in Computer and Electronic Engineering at North West University and works as a software developer at MUST Deep Learning.

  6. As of October 2026

    Cattle identification with deep vision

    He is a PhD student at North-West University, based at the MUST Research Office in Hermanus, where his doctoral work since 2025 applies deep vision models to the agriculture sector for real-time identification of individual cattle.

Career

  1. Since 2025MUST Deep LearningSoftware Developer

Education

  1. From 2025North-West UniversityPhD
  2. 2024North-West UniversityMaster of Computer and Electronic Engineering

From the biography.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes North-West University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  2. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI each year from 1955 to 2025, tracing the shift from university labs to companies.
  3. The United States employs an estimated 48% of people shaping AIWe estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  4. 43% of living people shaping AI work outside their country of birthWe estimate that 43% of living people shaping AI work outside their country of birth (520 of 1,198 people with known birth and employment countries).

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Harmen Potgieter yet. Send us a free-licensed photo.

  1. Hire Harmen P. | Kolabtree kolabtree.com. Checked 22 October 2026.
  2. Harmen Potgieter | CAIR cair.org.za. Checked 22 October 2026.
  3. [2501.14441] Impact of Batch Normalization on Convolutional Network Representations arxiv.org. Checked 22 October 2026.
  4. Publications | engineering.nwu.ac.za engineering.nwu.ac.za. Checked 22 October 2026.
  5. [2411.04695] Is network fragmentation a useful complexity measure? arxiv.org. Checked 22 October 2026.
  6. Harmen Potgieter | engineering.nwu.ac.za engineering.nwu.ac.za. Checked 22 October 2026.
  7. Meet the Team | engineering.nwu.ac.za engineering.nwu.ac.za. Checked 22 October 2026.