Track This Job
Add this job to your tracking list to:
- Monitor application status and updates
- Change status (Applied, Interview, Offer, etc.)
- Add personal notes and comments
- Set reminders for follow-ups
- Track your entire application journey
Save This Job
Add this job to your saved collection to:
- Access easily from your saved jobs dashboard
- Review job details later without searching again
- Compare with other saved opportunities
- Keep a collection of interesting positions
- Receive notifications about saved jobs before they expire
AI-Powered Job Summary
Get a concise overview of key job requirements, responsibilities, and qualifications in seconds.
Pro Tip: Use this feature to quickly decide if a job matches your skills before reading the full description.
The thesis will be carried out within the Advanced Geospatial AI team and will focus on the design of scalable geospatial data pipelines and the development of Vision Transformer (ViT)-based models for the semantic analysis of satellite imagery. The selected candidate will also contribute to the development of end‑to‑end GeoAI workflows, including large‑scale dataset ingestion and governance, fine‑tuning of foundation models (e.g., DINOv3), multi‑task architectures for building footprint segmentation and height estimation, as well as topology‑aware post‑processing pipelines for vector data generation.
The research will support industrial applications such as terrain database generation for flight simulators, UAV mission‑planning platforms, urban digital twins, and urban analytics systems, integrating data engineering, deep learning, computational geometry, and remote sensing within a real-world production environment.
Main responsibilities:
- Design and implement scalable pipelines for satellite imagery ingestion and preprocessing;
- Fine‑tune and validate Vision Transformer–based semantic segmentation models;
- Develop multi‑task learning architectures for building footprint segmentation and height estimation;
- Implement topology‑aware post‑processing algorithms for raster‑to‑vector conversion;
- Perform model evaluation, benchmarking, and validation across heterogeneous geographic scenarios;
- Collaborate with the R&D team to integrate AI models into industrial geospatial platforms.
- Knowledge of Python;
- Knowledge of of Deep Learning frameworks (PyTorch or TensorFlow);
- Fundamentals of Computer Vision;
- Understanding of semantic segmentation architectures.
- Basic knowledge of geospatial data formats (GeoTIFF, Shapefile, raster/vector);
- Experience with libraries such as GDAL, Rasterio, GeoPandas or equivalent;
- Experience with Vision Transformers and foundation models (DINO, MAE, etc.;
- Knowledge of multi‑task learning or regression models;
- Computational geometry and graph theory;
- Experience with PostGIS or spatial databases;
- Familiarity with MLOps and dataset governance practices;
- Experience with terrain databases or flight simulation environments.
Education: Bachelor’s or Master’s degree in STEM disciplines, preferably in Computer Scienceor computer engineering.
Why Choose TXT?
- Career opportunities in a rapidly growing and profoundly changing company with young, international staff;
- Training on business-related topics;
- Corporate Benefits (Ticket Restaurant, discounts as a group employee).
- Teamworking: Opportunity to collaborate with highly talented and passionate people in a highly professional development process;
- Hybrid work mode.
Key Skills
Ranked by relevanceReady to apply?
Join TXT GROUP and take your career to the next level!
Application takes less than 5 minutes

