Jakob Troidl

profile-picture.png

I am an AI Scientist at HHMI Janelia, part of the AI@HHMI initiative, where I lead the MIA-AI project. We build large-scale datasets and pre-trained models for microscopy image analysis, using self-supervised and multi-modal learning and architectures that reason across spatial scales.

My background is in computer science and connectomics, the reconstruction and analysis of neural circuits from nanometer-resolution microscopy data. I’ve worked on learned models of neuron morphology and self-supervised isotropic volume reconstruction, proofreading and annotation infrastructure (CAVE, SynAnno), and tools for analyzing circuit motifs (Vimo, MoMo). I also contributed to the largest human connectome to date and a connectome of the larval zebrafish brain.

I received my Ph.D. in Computer Science from Harvard, advised by Hanspeter Pfister, with research stays in the Turaga Lab at Janelia and at E11.bio as an ML Fellow. I hold a B.Sc. and M.Sc. in computer science from TU Wien.

news

Mar 04, 2026 🇸🇪 Keynote at the BIC workshop at ECCV in Malmö: Probing Scaling Laws in Microscopy Image Analysis.
Mar 04, 2026 💬 Attending AI and Biology at EMBL and the Synapses at Scale workshop at COSYNE.
Mar 04, 2026 📖 My PhD thesis, Scaling Computational Connectomics, is online.
Jan 26, 2026 📢 Started as an AI Scientist at HHMI Janelia.
Nov 18, 2025 🎓 Defended my PhD (recording).
Jul 25, 2025 📢 Three papers accepted to IEEE VIS 2025 and one to MICCAI EMA!
Jun 06, 2025 💬 Presented neuron shape reasoning at Analysis and Modeling of Connectomes and Frontiers in NeuroAI.
Apr 10, 2025 📝 CAVE is out in Nature Methods.
Mar 14, 2025 🔬 Joining E11.bio this summer as an ML Fellow on automated connectome proofreading.
Dec 21, 2024 💬 Talks on automated connectome proofreading at E11.bio, TU Wien, and Google Research.
Nov 24, 2024 📝 New preprint on automated connectome proofreading and typing with global neuron shape reasoning.
Oct 15, 2024 💬 Presented at the IEEE VIS Doctoral Colloquium (remote).
Jun 25, 2024 💬 Talk on neural implicit volume reconstruction at IST Austria.
Jun 23, 2024 💬 Talk on interactive motif analysis at the International Connectomics Conference in Berlin.
May 15, 2024 🧠 Started as a visiting researcher in the Turaga Lab at HHMI Janelia.
May 01, 2024 📝 New preprint on neural implicit isotropic volume reconstruction.
Feb 02, 2024 💬 Talk on human-centered computing in connectomics at Harvard Medical School.
Aug 02, 2023 📝 Vimo accepted to IEEE VIS 2023 🇦🇺.
May 21, 2022 📝 Two papers at EuroVis 2022: Barrio and a connectomics survey.
Jul 08, 2021 🎓 M.Sc. in Visual Computing from TU Wien (thesis).
Apr 15, 2021 🎉 Joining the Visual Computing Group at Harvard this fall for my Ph.D.
Aug 23, 2020 🏆 Best SciVis Paper at IEEE VIS 2020.
Jan 21, 2020 🎓 Among the first graduates of the TU Wien Informatics Bachelor with Honors program.

selected publications

  1. momo.png
    MoMo-Combining Neuron Morphology and Connectivity for Interactive Motif Analysis in Connectomes
    Michael Fekadu Shewarega* , Jakob Troidl*, Oliver Alvarado Rodriguez , and 6 more authors
    IEEE Transactions on Visualization on Computer Graphics , 2025
  2. gnsr.png
    Global Neuron Shape Reasoning with Point Affinity Transformers
    Jakob Troidl, Johannes Knittel , Wanhua Li , and 3 more authors
    bioarxiv, 2025
  3. niiv.png
    niiv: Interactive Self-supervised Neural Implicit Isotropic Volume Reconstruction
    Jakob Troidl, Yiqing Liang , Johanna Beyer , and 5 more authors
    MICCAI Workshop on Efficient Medical AI, 2025
  4. h01.jpeg
    A petavoxel fragment of human cerebral cortex reconstructed at nanoscale resolution
    Alexander Shapson-Coe , Michał Januszewski , Daniel R. Berger , and 28 more authors
    Science, 2024
  5. vimo-teaser-02.png
    Vimo: Visual Analysis of Neuronal Connectivity Motifs
    Jakob Troidl, Simon Warchol , Jinhan Choi , and 8 more authors
    IEEE Transactions on Visualization and Computer Graphics (IEEE VIS), 2023