AI Engineer - AI+CryoET

Howard Hughes Medical Institute (HHMI); Ashburn (VA); On-site; USD 90,000 - 120,000; Full time

Benefits offered by this job

Competitive compensation package

Comprehensive health and welfare benefits

On-site childcare

Access to world-class computational infrastructure

Partnership with AI labs

Job summary

Howard Hughes Medical Institute (HHMI) in Ashburn, Virginia, is seeking a specialist to join the AI+CryoET project, focusing on AI methods for particle detection in cryo-electron tomography. The role covers model development for gold nanoparticle detection and enhancing tomogram reconstructions.

Ideal candidates will have advanced degrees in relevant fields, extensive experience in deep learning, and excellent communication skills for effective interdisciplinary teamwork. The position offers competitive compensation and numerous benefits, including access to comprehensive computational resources.

Qualifications

• 3+ years training and evaluating deep-learning models on 3D or volumetric data.
• Experience with imaging detection, segmentation, or inverse problems preferred.
• Commitment to open science and collaborative work.

Responsibilities

• Develop and evaluate deep-learning models for particle detection in cryoET data.
• Design methods to improve tomogram reconstruction using gold-nanoparticle detections.
• Collaborate with interdisciplinary teams across institutions.

Skills

Deep-learning model training

Python programming

Collaborative communication

Education

Master’s or PhD in relevant field

Tools

PyTorch

JAX

Job description

About the Role

The position is part of the AI+CryoET project at HHMI, focused on developing AI methods for particle detection and structural analysis in cryo-electron tomography (cryoET) data. The role involves collaborating with experimental and computational scientists at several institutions to create supervised and self-supervised model architectures that can detect gold‑nanoparticle probes, identify nucleosome arrangements, and improve tomogram reconstructions.

Responsibilities

• Develop and evaluate deep‑learning models for detecting and localizing gold nanoparticles and macromolecular particles (e.g., nucleosomes, synaptic receptors) in cryoET data.• Design methods that use gold‑nanoparticle detections to improve tomogram reconstruction, addressing challenges such as tilt‑series alignment, deformations, and low signal‑to‑noise conditions.• Build rigorous AI training and evaluation pipelines, including handling of missing‑wedge artifacts, CTF effects, and sim‑to‑real transfer from molecular‑dynamics‑derived synthetic training data.• Identify where additional human annotation and proofreading will be most helpful and guide annotation efforts.• Contribute to scientific publications, present findings at conferences, and maintain a well‑documented codebase that enables reproducibility and extension of results.• Collaborate with interdisciplinary teams across multiple institutions.

Qualifications

• Master’s or PhD in Computer Science, Applied Mathematics, Physics, Computational Chemistry, or a related field, or an equivalent combination of education and experience.
• 3+ years training and evaluating deep‑learning models, especially on 3D or volumetric image data.
• Experience with detection, segmentation, or inverse problems in imaging is strongly preferred.
• Strong Python skills and proficiency in PyTorch and/or JAX.
• Ability to reason about neural‑network behavior from first principles: how architectural choices, regularization, and training procedures affect model behavior.
• Rigorous experimental design skills (model comparisons, ablation studies, reproducibility).
• Commitment to open science.
• Experience with scalable GPU‑based computing environments on Linux HPC clusters and high‑throughput processing for large‑scale data.
• Excellent communication skills and interest in interdisciplinary collaboration.
• Optional: experience with cryo‑EM/ET data processing, tomographic reconstruction, or related inverse problems; familiarity with molecular‑dynamics simulations (OpenMM, LAMMPS); knowledge of cryoET software tools (IMOD, Warp, RELION, AreTomo) or file formats (MRC, Zarr); experience with template matching or sub‑tomogram averaging; familiarity with differentiable rendering or neural radiance fields.

Benefits

• Competitive compensation package with comprehensive health and welfare benefits.
• Supportive team environment that promotes collaboration and knowledge sharing.
• Access to world‑class computational infrastructure, GPU‑based computing environments, and unique high‑quality cryoET datasets.
• Opportunities to work directly with leading structural biologists, cryoET experimentalists, and molecular‑dynamics experts on highly interdisciplinary projects.
• Work‑life balance amenities such as on‑site childcare, free gyms, on‑campus housing, social and dining spaces, and a shuttle bus service to Janelia from the Washington, D.C. metro area.
• Partnership with frontier AI labs on scientific applications of AI.

Equal Opportunity Employer

HHMI is an Equal Opportunity Employer. We employ a rigorous process to evaluate and provide reasonable accommodations for all applicants.