Machine Learning Engineer, Connectomics

Eon Systems•San Francisco, CA

33d

About The Position

Eon is building the infrastructure for large-scale connectomics data collection, reconstruction, and brain simulation. Our mission is to enable the safe and scalable development of brain emulation technology, beginning with digital twins of model organisms. We are developing an end-to-end platform that spans tissue preparation, high-throughput microscopy, large-scale image processing, neural reconstruction, connectome-based modeling, and embodied simulation. We are looking for exceptional engineers and scientists who can help turn biological brain data into usable computational systems. We are seeking a machine learning, software, or data engineer with strong experience in large-scale neuroscience data pipelines. The ideal candidate has worked with connectomics, volumetric imaging, segmentation workflows, manual or semi-automated proofreading pipelines, and large-scale n-dimensional image data. This role will help build and optimize Eon’s connectomics reconstruction pipeline: from raw microscopy data to segmented neurons, synapses, connectivity maps, visualizations, and brain simulations. You will work on segmentation, affinity prediction, watershed/post-processing, data management, scalable visualization, and machine-learning experiments. You may also contribute to embodied simulations of animal models using connectome-derived neural architectures. This is a hands-on role for someone who is comfortable moving between ML experimentation, production data infrastructure, scientific computing, and computational neuroscience.

Requirements

Strong ability to create polished and engaging visualizations.
Neuroglancer, BigDataViewer, Fiji/ImageJ, CloudVolume, TensorStore, Zarr, N5, DVID, CAVE, or related tools.
Affinity prediction, watershed segmentation, flood filling networks, U-Nets, transformers for vision, or other computer vision models for biological image data.
Distributed data processing, cloud infrastructure, GPU inference, and high-throughput ML pipelines.
Large-scale n-dimensional array processing in Python, C++, Java, or similar environments.
Strong software engineering skills, including clean code, version control, testing, documentation, and reproducible workflows.
Experience with large data systems, ideally at TB scale or above.
Experience with computer vision, biological image segmentation, or volumetric data analysis.
Strong communication skills and ability to collaborate with neuroscientists, microscopists, ML engineers, and data infrastructure engineers.

Nice To Haves

GPU kernel development experience is a definite plus.

Responsibilities

Build, optimize, and maintain large-scale connectomics data pipelines for volumetric microscopy data.
Develop and improve machine learning workflows for image segmentation, affinity prediction, watershed/post-processing, synapse detection, and neural reconstruction.
Work with large-scale n-dimensional image data, including TB- to PB-scale datasets.
Run controlled ML experiments to improve segmentation accuracy, throughput, and reliability.
Create polished, compelling visualizations of connectomic data, neural activity, and reconstructed circuits.