"COLCOM - Multimodal Crop Analysis & Fertilizer Optimization for Sustainable Agriculture"

Job Details

About the recruiter — UM6P

Mohammed VI Polytechnic University (UM6P) is a research-and-innovation focused university in Morocco committed to African development. UM6P’s College of Computing (Benguerir & Rabat campuses) advances world-class research and education in Computer Science, fostering partnerships with industry and local stakeholders.

Project summary

Develop farmer-centric systems that fuse multi-modal remote sensing, soil and phenology data to enable crop classification and precise, customized fertilizer recommendations.

Selection criteria

Required

PhD (awarded or defended before start) in Computer Science, Remote-Sensing/Geoinformatics, Agricultural Data Science, or related field.

Strong track record in remote-sensing imagery and/or time-series analysis and ML/DL for spatio-temporal data.

Advanced Python skills and experience with ML frameworks and geospatial tools (e.g., PyTorch/TensorFlow, rasterio/GDAL).

Ability to work independently and produce reproducible research outputs.

Good English (written & oral) and willingness to collaborate with agronomists and partners.

Preferred

Postdoc or ≥2 years research experience after PhD; first-author publications in relevant journals/conferences.

Experience with multimodal data fusion (optical/SAR/soil/phenology), satellite platforms (Sentinel/Landsat/GEE), and building reproducible pipelines.

Field/ground-truth experience, agronomic knowledge, or fertilizer-recommendation systems. French/Arabic useful for local engagement.

Application materials (required)

1. Cover letter (fit with CropID + available start date).

2. CV with links (ORCID, GitHub).

3. Research statement (1–2 pages) with a 12–18 month plan.

4. Up to 3 representative papers and links to code/datasets (if available).

5. 2–3 referee contacts.

Selection & timeline

Shortlist based on research fit, technical skills, and interdisciplinarity. Top candidates invited for a technical interview covering past projects, a 6-month plan, reproducibility practices, and farmer-translation. Appointment: fixed-term (24 months), UM6P (Benguerir).

References

Moreno-Revelo, M.Y., et al. (2021). Enhanced convolutional-neural-network architecture for crop classification. Applied Sciences, 11(9), 4292.

Bhattacharya, S. & Pandey, M. (2024). PCFRIMDS: Smart Next-Generation Approach for Precision Crop and Fertilizer Recommendations Using Integrated Multimodal Data Fusion for Sustainable Agriculture. IEEE Transactions on Consumer Electronics.