Telecom Mediation Engineer
Minor Hotels Europe and Americas
AnsibleDockerJavaJenkinsKubernetesOraclePerlPostgresPythonSQLUnix
ML Engineer, Computer Vision – Geospatial & Environmental Analysis
Helpware Tech
full-time
Posted on:
Location: 🇺🇦 Ukraine
Visit company websiteJob Level
Mid-LevelSenior
Tech Stack
PythonPyTorch
About the role
- Contribute to geospatial and environmental risk analysis using high-resolution drone imagery
- Fine-tuning and benchmarking zero-shot object detection and depth estimation models (e.g., Grounding DINO, OV-DINO)
- Incorporate DSM/DTM data layers to improve contextual depth estimation
- Develop reproducible evaluation notebooks and visualizations, mapping outputs to wildfire resilience standards
- Collaborate with domain experts in geospatial analytics and wildfire risk to ensure models meet real-world criteria
- Transition from open-vocabulary to closed-set detection as labeled datasets become available
- Ensure high standards in code quality, version control, and experiment reproducibility
Requirements
- Proven experience in deep learning for computer vision, especially object detection and depth estimation
- Hands-on familiarity with zero-shot detection frameworks (Grounding DINO, OV-DINO, etc.)
- Knowledge of geospatial data and DSM/DTM modeling
- Strong programming skills in Python, with experience using PyTorch, OpenCV, torchvision, or similar libraries
- Experience building and maintaining Jupyter-based evaluation workflows
- Solid data visualization skills (e.g., using Matplotlib, Plotly, etc.)
