Benedikt Blumenstiel
AI for Climate Impact, IBM Research | Based in Zurich
About me
Hello everyone! I'm Benedikt — an open-minded researcher passionate about machine learning, innovation, and sustainability. I work in the AI for Climate Impact Team at the IBM Research Zurich lab. We build foundation models for earth observation and work on climate-related applications using satellite data. Eager to share ideas and experiences in the intersection of machine learning and sustainability.
Looking forward to connecting with you!
Selected projects
Prithvi EO 2.0
Prithvi-EO-2.0 is the second generation EO foundation model jointly developed by IBM, NASA, and Jülich Supercomputing Centre. Prithvi-EO-2.0 is based on the ViT architecture, pretrained using a masked autoencoder (MAE) approach with two modifications: 3D positional embeddings for multi-temporal input and additional location and temporal embeddings to encode metadata. The models are pre-trained on 4.2M time series samples and released on HuggingFace.
Remote Sensing Image Retrieval
Embeddings of Geospatial Foundation Models (GeoFMs) enable simple but accurate content-based image retrieval of remote sensing images. We tested the GeoFM Prithvi (IBM-NASA) on two challenging datasets with multi-spectral satellite images and outperformed common computer vision models by a large margin. Check out our repository for more details.
MESS Benchmark
Holistic benchmarks allow a robust quantitative comparison of model architectures. However, such evaluation practice covering a wide range of real-world applications is lacking for multi-modal (i.e., text-to-image) zero-shot semantic segmentation tasks. Thus, our motivation to compose the Multi-domain Evaluation of Semantic Segmentation (MESS) benchmark, which consists of 22 datasets for varying application domains. More details can be found on our website.
CLAIMED Component Compiler (C3)
The CLAIMED framework enables rapid development and deployment of processing steps, e.g. for data preprocessing in deep learning. Using the CLAIMED Component Compiler (C3), components can be easily built from a Python script or notebook and deployed on KubeFlow, Kubernetes, or locally via CLI. Notably, C3 eliminates the need for Docker or Kubernetes expertise, making it accessible to any data scientist to deploy code on clusters. Check out the Getting Started page for more information.
Résumé
Work Experience
Research Software Engineer, AI for Climate Impact, IBM Research
02/2024 - present
- Research on pre-training of multi-satellite and multi-modal geospatial foundation models
- Developing the data pipeline for a geospatial ML platform and foundation model pre-training
Intern, AI for Climate Impact, IBM Research
08/2023 - 02/2024
- Deploying and applying MLOps tools and pipelines
- Developing the CLAIMED Component Compiler (C3) for the CLAIMED project
- Experiments with remote sensing image retrieval using geospatial foundation models
Junior Researcher, Applied AI in Services Lab, KIT
05/2020 - 04/2022
- Research about human-AI collaboration and machine learning, with a focus on few-shot learning
- Developed AI applications for a AI project in the AEC industry
- Worked with multiple computer vision models and deployed them on high performance clusters
Design Thinking project, KIT & HUK Coburg
10/2019 - 06/2020
- Used design thinking to develop user-centered innovations and prototypes for retirement saving
Intern, R&D Strategy, AUDI AG
07/2017 - 02/2018
- Supported the project manager in defining a new technology-focused R&D strategy
- Worked in the project management office of an R&D transformation project
Education
M.Sc. Information Systems, Karlsruhe Institute of Technology
10/2019 - 04/2023
- Focus on machine learning and service innovation
Exchange, Norwegian University of Science and Technology
01/2021 - 06/2021
- Courses in computer vision, natural language processing and statistics
B.Sc. Industrial Engineering, Karlsruhe Institute of Technology
10/2015 - 09/2019
- Focus on digital services and business strategy