Welcome, I'm Magdalena

AI@HHMI | AI × Science | AI × Imaging

I'm an independent Machine Learning Researcher in the AI@HHMI initiative at HHMI Janelia Research Campus, leading research projects in AI for Science at the intersection of machine learning, microscopy, physics, and cell biology — spanning AI methods for cryo-electron tomography, agentic AI for microscopy, and AI methods for adaptive optics. I work closely with experimentalists and computer scientists across fields.

Portrait of Magdalena C. Schneider

News

Research Interests

  • Cryo-electron tomography
  • Agentic AI for microscopy
  • Computational adaptive optics
  • Physics-based simulations
  • Implicit neural representations
  • Generative models
  • Graph neural networks
  • Differentiable physics

Projects

AI+CryoET thumbnail

AI+CryoET

Combining state-of-the-art techniques in cryo-electron tomography with tailored AI methods to reveal cellular architecture and function at molecular scale. Part of the AI@HHMI initiative.

Collaboration with HHMI Investigators Mike Rosen (UTSW), Eric Gouaux (OHSU), and Elizabeth Villa (UCSD), as well as Rosana Collepardo-Guevara (Cambridge). I lead AI method development at Janelia.

Agentic microscopy thumbnail

Agentic microscopy

Building AI agents that autonomously operate microscopes. Developed on light-sheet microscopes built by the Shroff lab, with C. elegans developmental imaging as the driving use case.

Jointly led with Hari Shroff (Janelia), developed with P. S. Kesavan, in collaboration with Anthropic.

Computational adaptive optics thumbnail

Computational adaptive optics

Correcting sample-induced aberrations in fluorescence microscopy through computational wavefront sensing and image reconstruction, spanning 2D and 3D imaging across multiple microscopy modalities.

Collaboration with the Shroff lab (Janelia).

Optical simulation thumbnail

Optical simulation

Physics-based forward models of optical systems, enabling accurate PSF fitting, image reconstruction, and optical system design.

I developed a PSF simulation framework for single-molecule localization microscopy and contributed to Chromatix, a differentiable GPU-accelerated wave-optics library.

Selected Publications

  1. DeepPD: Joint phase and object estimation from phase diversity with neural calibration of a deformable mirror

    Preprint (arXiv) · 2025

  2. Chromatix: a differentiable, GPU-accelerated wave-optics library

    bioRxiv · 2025

    Preprint

  3. Phase-diversity-based wavefront sensing for fluorescence microscopy

    Optica · 2024

  4. Interactive simulation and visualization of point spread functions in single molecule imaging

    Optics Communications · 2024

Full list: Google Scholar · ORCID

Blog Contributions

Posts I've co-authored on methods and advances in AI and machine learning:

HHMI Janelia Research Campus