Skip to content
@HelmholtzAI-Consultants-Munich

HelmholtzAI-consultants-Munich

Leading in applied AI by combining unique research questions, data and expertise with democratized access to new tools.

Helmholtz AI Consultants @ Helmholtz Munich

Website Email License


Who we are

We are a health-focused AI consultancy team based at Helmholtz Munich, operating as part of Helmholtz AI - Germany's national AI initiative for science.

Our team supports life science researchers in applying machine learning and deep learning to their research - from early project scoping through to implementation, code review, and publication. We work across imaging, omics, clinical data, time series, and responsible AI, collaborating with research groups across the Helmholtz Association.


What we do

We are key actors in Helmholtz AI's mission of democratising AI for data-driven science.

  • AI consulting - we advise and co-develop ML solutions for research projects, at any stage from idea to working pipeline
  • Pair programming & code review - hands-on support to help scientific teams build reproducible, maintainable code
  • Open-source tools - we develop and maintain reusable software for the research community
  • Teaching - we design and deliver courses on machine learning, explainability, and LLMs for scientists

Selected repositories

Research & Tools

Repository Description
oligo-designer-toolsuite Modular Python framework for custom oligo probe design pipelines, including the CycleHCR probe-design pipeline (spatial transcriptomics, FISH, oligo-seq)
fg-clustering Forest-Guided Clustering - computes feature importance from decision-path subgroups within Random Forest models, avoiding misleading correlated-feature interpretations
data-centric-platform BATS: data-centric annotation platform for microscopy images, using baseline models (e.g. CellPose) plus expert-in-the-loop correction to minimise labelling time

We have many more projects than these three - check out the rest of our repositories here: → Browse all repositories

Teaching materials

Repository Description
XAI-Tutorials Self-contained tutorials on model-agnostic and model-specific explainability methods
DimRed-Course 4-hour hands-on introduction to dimensionality reduction for high-dimensional biological data
llm-transformers-course Lecture notes, code, and exercises on Transformers and Large Language Models

Work with us

Research groups within the Helmholtz Association can request AI consulting support through the Helmholtz AI voucher system:

→ Request a consultation

We offer Exploration vouchers (short, scoping-focused) and Realization vouchers (longer implementation projects). If you are unsure whether your project is a good fit, feel free to reach out by email first.

→ Contact us

Pinned Loading

  1. Automatic-Cell-Counter Automatic-Cell-Counter Public

    A GUI based on napari for counting cells automatically in cell migration experiments

    Python 47 10

  2. Quicksetup-ai Quicksetup-ai Public

    A flexible template, as a quick setup for deep learning projects in pytorch-lightning

    Python 48 7

  3. PySDDR PySDDR Public

    A python package for semi-structured deep distributional regression

    Jupyter Notebook 23 10

  4. fg-clustering fg-clustering Public

    Explainability for Random Forest Models.

    Jupyter Notebook 34 14

  5. oligo-designer-toolsuite oligo-designer-toolsuite Public

    Develop custom oligo design pipelines.

    Python 29 11

  6. napari-organoid-counter napari-organoid-counter Public

    A napari plugin for counting organoids from brightfield microscopy images with deep learning

    Python 15 3

Repositories

Showing 10 of 44 repositories

Top languages

Loading…

Most used topics

Loading…