Job Summary
Position Overview Seeking an undergraduate student to support development and testing of AI and image-processing methods for engineering prototypes. The student will work with images, data, probability-based models, and machine-learning tools to help detect, classify, and evaluate physical system states.
What You Will Do
● Develop and test image-processing and computer-vision methods using Python.
● Work with camera images to identify objects, connections, patterns, and incorrect configurations.
● Prepare datasets, label images, extract features, and evaluate model performance.
● Experiment with classical computer vision and machine-learning approaches.
● Analyze uncertainty and probability in detection and classification results.
● Document experiments, results, and technical decisions.
Who Should Apply
Purdue juniors or seniors in Computer Science, Electrical/Computer Engineering, Data Science, Mathematics, Statistics, Mechanical Engineering, or a related field. You do not need to know every topic listed above. Strong mathematical reasoning, curiosity, and willingness to learn matter most.
Education
Experience
Useful Background
● Python programming and comfort working with data.
● Image processing or computer vision, such as OpenCV, filtering, segmentation, feature extraction, or object detection.
● Probability and statistics, including random variables, distributions, conditional probability, Bayes' rule, expectation, variance, and Markov's inequality.
● Linear algebra, including vectors, matrices, transformations, and eigenvalues/eigenvectors.
● Calculus and basic optimization concepts.
● Interest in stochastic processes and Markov chains, including the Markov property, state transitions, and transition probabilities.
● Machine-learning fundamentals such as classification, training/testing data, loss functions, and model evaluation.
● Interest and eagerness to learn and work with a diverse team, including virtual work with collaborators in different industry sectors and time zones, building a hands-on kit for a variety of communities of learners.
Helpful, But Not Required
● PyTorch, TensorFlow, scikit-learn, NumPy, or pandas.
● Convolutional neural networks, object detection, or image classification.
● Experience with cameras, embedded systems, robotics, or engineering prototypes.
● Coursework in AI, machine learning, computer vision, probability, statistics, signals, or applied mathematics.