We are seeking a highly specialized ML/AI Data Scientist to develop and train a machine learning model for our AI-powered duck tracking API. The core of this project is to analyze NEXRAD weather radar data to identify and classify waterfowl roosting signatures. You will be responsible for the entire machine learning lifecycle, from data preparation and model training to evaluation and deployment. This is a unique opportunity to work on a cutting-edge project at the intersection of machine learning, computer vision, and environmental science. You will collaborate with a Cloud Infrastructure Engineer who will be building the data pipeline and API. Your primary focus will be on developing a highly accurate model for roost detection. You will be provided with a comprehensive technical guide that outlines the system architecture and implementation steps. Project Overview The goal is to create a machine learning model that can: 1. Process and prepare radar imagery for training. 2. Accurately detect and classify bird roost signatures from other atmospheric phenomena (weather, insects, etc.). 3. Be deployed in a cloud environment for real-time or batch inference. Key Deliverables • A well-documented and reproducible machine learning pipeline for data preparation and training. • A trained, evaluated, and production-ready Convolutional Neural Network (CNN) for roost detection. • Clear documentation of the model architecture, training process, and performance metrics. Responsibilities • Collect, label, and augment a dataset of radar images for training and validation. • Design, build, and train a Convolutional Neural Network (CNN) using TensorFlow or PyTorch to classify bird roosts. • Experiment with different model architectures and hyperparameters to optimize performance. • Evaluate the model's accuracy, precision, and recall, and iterate as needed. • Package and deploy the trained model for use in the AWS data pipeline (e.g., as a SageMaker endpoint). • Collaborate with the Cloud Infrastructure Engineer to ensure seamless integration of the model into the overall system. Required Experience & Skills • Proven experience as a Data Scientist or Machine Learning Engineer with a strong portfolio of successfully deployed models. • Expertise in Python and its data science ecosystem (NumPy, Pandas, Scikit-learn). • Deep knowledge of machine learning, particularly with deep learning frameworks like TensorFlow or PyTorch. • Hands-on experience building and training Convolutional Neural Networks (CNNs) for image classification tasks. • Familiarity with data processing and analysis of large-scale datasets. • Experience with the full machine learning lifecycle, from data collection to model deployment. • Excellent problem-solving skills and the ability to work independently. Preferred Qualifications • Experience with AWS SageMaker or other cloud-based machine learning platforms. • Familiarity with radar data or other forms of remote sensing imagery. • Experience with data augmentation and transfer learning techniques. • A background in computer vision or a related field. How to Apply Please submit your proposal with the following: 1. A cover letter detailing your experience with the required skills and technologies. 2. Links to your portfolio or examples of similar machine learning projects you have completed. 3. Your estimated timeline and budget for this project. 4. Please start your application with the phrase "NEXRAD ML Specialist" to show you have read the entire description. We have a detailed technical guide that will be shared with shortlisted candidates. We look forward to seeing your proposals! Apply tot his job