The primary responsibilities of this role are to:

Solve real agricultural problems using deep learning and AI across image and other data modalities, translating complex models into tangible business and scientific impact.

Design and implement end-to-end machine learning pipelines for computer vision use cases, including segmentation, classification, detection, and multi-task learning.

Prototype, evaluate, and iterate on cutting-edge architectures such as CNNs, Vision Transformers, foundational and large-scale vision models, ensuring state-of-the-art performance.

Optimize models for accuracy, robustness, and inference efficiency, including experimentation with hyperparameters, compression, and deployment-oriented optimisations.

Independently build scalable data pipelines for training, validation, and evaluation, including data ingestion, augmentation strategies, and active learning loops.

Collaborate cross-functionally with product, data, and software engineering teams to integrate models into production systems and deliver reliable, maintainable solutions.

Contribute to MLOps practices, including model versioning, deployment, monitoring, and retraining workflows using modern tooling and cloud-based platforms.

Build strong cross-functional relationships and actively engage with the broader Data Science Community to share best practices, align on standards, and co-create innovative solutions.

Present clear, compelling, and validated stories about experiments, results, and recommendations to peers, senior management, and internal customers to drive strategic and operational decisions.

We seek an incumbent who possesses the following:

M.S. with 2+ years of experience or Ph.D. in Computer Science, Electrical Engineering, or a related field with a focus on machine learning or computer vision.

Proficiency in Python and experience with deep learning frameworks such as PyTorch or TensorFlow.

Hands-on experience with modern computer vision architectures including models such as ResNet, UNet, DeepLab, YOLO, SegFormer, SAM, and Vision Transformers.

Strong background in handling large-scale datasets and creating custom datasets, for example using frameworks such as Hugging Face Datasets.

Solid understanding of core machine learning concepts including loss functions, regularization, optimisation, and learning rate scheduling.

Experience developing and deploying models using cloud-based ML platforms such as AWS SageMaker.

Familiarity with Unix environments, including bash, file systems, and core utilities.

Strong engineering practices including use of Git, Docker, CI/CD pipelines, modular codebase design, and unit testing.

XML job scraping automation by YubHub