AI/ML Engineering Intern — Medical Imaging

neuro42 is a Bay Area medical technology company building medical imaging and surgical robotics systems. As an AI/ML Engineering Intern, you will help improve the images from our portable low-field MRI scanner by combining them with a patient's earlier 3T scan. You will work closely with the imaging and software teams on image registration, image enhancement, evaluation, and the tooling that supports all of it.

You will see the full cycle of how an algorithm becomes part of a medical device: preparing data, building the method, measuring how well it works, testing it, and writing it up. We are looking for an undergraduate or graduate student with strong Python skills, a good grasp of image processing and linear algebra, real care about getting things right, and an interest in medical imaging and MRI.

Responsibilities may include, but are not limited to:

1 . Build and test methods that align a patient's earlier 3T scan with a new low-field scan, including rigid and deformable registration, similarity metrics that work across different image contrasts, and checks that catch when alignment has failed. 
2 . Prepare images acquired at different field strengths so they can be compared: normalizing intensities, correcting bias fields, resampling onto a common grid, and handling LPS and RAS orientation conventions correctly. 
3 .Implement and compare ways of using the 3T scan to sharpen and denoise the low-field image, covering both classical methods and deep learning. 
4 . Set up the infrastructure behind that work: organizing datasets, assembling matched case pairs, running experiments reproducibly, and tracking results. 
5 . Write tooling that scores the results on SNR, CNR, sharpness, and how well structures line up with a reference scan. 
6 .Build tests that check the enhanced image is faithful to the new scan. The earlier scan is months old, so we need to know the method is not adding detail that is no longer there. 
7 . Help measure geometric accuracy in low-field images, including B0 inhomogeneity and gradient nonlinearity, and confirm that enhancement does not hide errors that matter for image-guided procedures. 
8 . Package the work as a pipeline the team can run end to end on real scanner data, with unit and regression tests and CI support. 
9 .Build internal tools that run the pipeline across many cases and produce clear comparison reports for engineering and clinical review. 
10 .Document what you built, how you tested it, what you found, and what broke, and help track issues to closure. Keep project documentation organized, including experiment records, engineering notes, and issues in Jira. 
11 .Work with imaging, software, hardware, and robotics engineers during development and integration. 
12 .Share progress, results, and findings with the engineering team and, when appropriate, with company leadership.

Preferred Qualifications:

1 .Currently pursuing a Bachelor's or Master's degree in Computer Science, Electrical Engineering, Biomedical Engineering, Medical Physics, Applied Mathematics, or a related field. 
2 .Strong Python, including NumPy and a testing framework such as pytest. 
3 .Experience with PyTorch, including building and training a model yourself. 
4 .Linear algebra you can apply — coordinate frames, transformation matrices, least squares. 
5 .Solid basics in signal and image processing: convolution, filtering, sampling, and the Fourier transform. 
6. Familiarity with 3D medical image formats such as NIfTI or DICOM, and with at least one of SimpleITK, nibabel, ANTs, ITK-SNAP, or 3D Slicer. 
7. Exposure to image registration, inverse problems, or deep learning for denoising, super-resolution, or image-to-image translation is a plus. 
8 .Some familiarity with MRI physics or reconstruction, such as k-space, parallel imaging, or compressed sensing, is a plus. 
9 .Experience with version control and working on a shared codebase, including code review. 
10 . Familiarity with regulated software standards such as IEC 62304 or ISO 13485, or with de-identified clinical data, is a plus. 
11 .Interest in medical imaging, MRI, surgical robotics, or clinical software. 
12 .Strong problem-solving skills, attention to detail, and the habit of checking results with numbers rather than by eye. 
13 .Able to work well on your own and as part of a multidisciplinary team. 
14 .Strong written and verbal communication skills.

This internship requires at least 20 hours per week. You will get hands-on experience building and validating imaging and machine learning methods for a real medical device, and your work on the pipelines and tooling will be used by the engineering team day to day.