EngineerJobs.io
← Back to all jobs

Job Description

The Howard Hughes Medical Institute (HHMI) Janelia is seeking an AI Engineer to build and deploy production-ready pipelines for reconstructing, segmenting, and proofreading connectomes from expansion microscopy data. This role supports AI@HHMI’s Foundational Microscopy Image Analysis (MIA) initiative, with an emphasis on improving reconstruction quality and scaling throughput for large, compute-intensive workloads.

Working on the Janelia Research Campus in Ashburn, VA, you will take models and workflows from prototype to reliable production systems, ensuring they perform against defined reconstruction quality measures while integrating effectively with scientific data and tooling.

Role focus

  • Build and deploy production workflows for connectome reconstruction, segmentation, and proofreading from expansion microscopy (ExM) data.
  • Scale inference and end-to-end reconstruction pipelines to large GPU cluster workloads.
  • Support training and evaluation as new ground-truth data becomes available, improving quality and throughput against agreed benchmarks.
  • Release pipelines, models, and reconstructions to the broader scientific community with attention to reproducibility and downstream usability.

Responsibilities

  • Reconstruct connectomes at scale using pre-trained segmentation models.
  • Train and fine-tune models as new ground-truth data is collected.
  • Design and run rigorous model evaluations to measure and improve reconstruction quality.
  • Optimize inference pipelines to increase reconstruction throughput on GPU clusters.
  • Build and support proofreading workflows, including integration with CAVE.
  • Coordinate manual annotation and proofreading to strengthen training and evaluation datasets.
  • Own and run production reconstruction pipelines end-to-end on new ExM datasets.
  • Demonstrably improve throughput or reconstruction accuracy relative to an agreed benchmark.
  • Deliver high-quality connectome segmentations that require minimal manual proofreading.
  • Advise stakeholders on data strategy and best practices, and potentially mentor junior engineers.

Requirements

  • 3+ years of training and evaluation experience with deep learning models (Transformers, U-Nets), ideally on 3D image data.
  • Experience deploying models with workflow orchestration and high-throughput processing on GPU clusters using tools such as Kubernetes, Nextflow, or LSF.
  • Ability to explain reasoning behind results when outcomes change.
  • Understanding of how architecture and regularization choices affect performance, including designing experiments to validate explanations.
  • Strong Python skills in PyTorch or JAX.
  • Familiarity with modern vision models such as DINOv3, SAM, Cellpose, and Vision Transformers.
  • Experience defining reconstruction quality measures first, then evaluating changes against that measure.
  • Interest in learning neuroscience and cell biology from practitioners and collaborating across disciplines.
  • Comfort working in a shared, git-based codebase.

Technologies

  • Python, PyTorch, JAX
  • Transformers, U-Nets
  • Kubernetes, Nextflow, LSF
  • DINOv3, SAM, Cellpose, Vision Transformers
  • CAVE, git

Compensation

USD 128,050 - 239,789 per year.

Education

Master’s degree or PhD (or an equivalent combination of education and experience).

Benefits

  • Competitive compensation package with comprehensive health and welfare benefits.
  • Supportive team environment promoting collaboration and knowledge sharing.
  • Access to world-class computational infrastructure and high-quality datasets.
  • Opportunity to work with leading researchers, software engineers, and AI/ML experts on fundamental scientific questions.
  • Amenities supporting work-life balance, including on-site childcare, free gyms, on-campus housing, social and dining spaces, and a shuttle bus service from the Washington, D.C. metro area.
  • Exceptional health benefits, retirement plans, time off, and recognition and wellness programs.

Nice to have

  • Microscopy image analysis experience, especially 3D electron or expansion microscopy.
  • Knowledge of connectome reconstruction methods such as Flood-Filling Networks (FFN), Local Shape Descriptors (LSDs), and graph agglomeration.
  • Experience with connectomics and microscopy tools such as CAVE, Neuroglancer, and Zarr.
  • Experience with self-supervised methods such as DINO or masked autoencoders.
  • Experience working with large biological datasets.

Practical details

  • On-site role at Janelia Research Campus in Ashburn, VA.

Physical requirements

  • Remain in a normal seated or standing position for extended periods.
  • Reach and grasp by extending hand(s) or arm(s); use keyboard and other tools requiring finger dexterity.
  • Communicate using the spoken word; see and hear within normal parameters.
  • Move about the workspace and handle materials weighing up to several pounds (for example, a laptop computer or tablet).
  • Reasonable accommodations may be available for qualified individuals with disabilities, evaluated on a case-by-case basis.

Similar Jobs