AI Engineer - Metabolic Sensor Design

Intro: 

AI@HHMI: HHMI is investing $500 million over the next 10 years to support AI-driven projects and to embed AI systems throughout every stage of the scientific process in labs across HHMI. The AI initiative will be centered at HHMI’s Janelia Research Campus. Janelia has been at the forefront of AI-driven research in biology for more than 15 years. Its forward-thinking structure, centralized funding, and collaborative culture make it ideally suited to take this bold leap forward. To learn more about the initiative, visit here.   

Please include a cover letter with your application detailing your qualifications and experience as they relate to this position. This should include a description of a deep learning project that you have executed, ideally a creative use of a transformer-based or related architecture that you trained yourself. If it is in the sequence or protein structure domain, even better! If possible, include a link to a code repository. If you are a contributor to a joint project, that is wonderful, but please describe specifics of your contribution to the project. Briefly discuss the results of the project as well as limitations and challenges you encountered.

Also, please include a link to your GitHub profile and/or links to relevant projects at the bottom of your cover letter.   

About the role: 

The CombinAItorial Sensor Design project is part of HHMI’s AI for Science Initiative (ai.hhmi.org) and brings together expertise in protein engineering, advanced microscopy, and machine learning. Our goal is to develop a protein biosensor optimization pipeline that integrates high-throughput functional screening with predictive deep learning to accelerate the directed evolution of protein biosensors for visualizing dynamic biochemical processes in living cells. 

We are seeking a highly skilled AI Software Engineer to join our team and play a crucial role in advancing our AI-driven scientific initiatives. In this position, you will be responsible for developing and maintaining the computational infrastructure essential for AI-powered biosensor optimization. You will collaborate with data scientists and experimentalists to develop robust data flows from optical pooled screening outputs through to model training and deployment. You will implement cutting-edge tools for predicting fluorescence properties and biochemical performance based on protein sequence and structure. You will utilize generative models to produce new sequences for biosensor candidates that will be tested in the lab. This role will require deep knowledge of the underlying models as well as practical implementation skills to have the maximum biological impact. You will lead comparative studies, implement novel architectures, and ensure all work meets the highest standards of reproducible, open science.  This role requires close collaboration with our microscopy, sequencing, and protein engineering team to ensure the seamless integration of computational and experimental workflows. 

Strong programming skills in Python, PyTorch, and JAX are required, along with the ability to reason about neural network behavior from first principles. We seek candidates who can think critically about model design, understand how architectural choices and regularization affect model behavior, and design rigorous experiments to evaluate models. Domain expertise in sequence or protein structure analysis will be highly valued. Because this is a team project, we value a clean shared codebase and git-based collaborative workflows. Familiarity with protein modeling or machine learning frameworks such as AlphaFold, ESM, Chai-1, or Boltz-1 is highly valued. We are looking for candidates with experience in ML model deployment, workflow orchestration, and high-throughput data processing, as well as experience working with large biological datasets in GPU-based computing environments. 

What we provide:

• A competitive compensation package, with comprehensive health and welfare benefits.

• A supportive team environment that promotes collaboration and knowledge sharing.

• The opportunity to engage with world-class researchers, software engineers and AI/ML experts,  contribute to impactful science, and be part of a dynamic community committed to advancing humanity’s understanding of fundamental scientific questions.

• Amenities that enhance work-life balance such as on-site childcare, free gyms, available on-campus housing, social and dining spaces, and convenient shuttle bus service to Janelia from the Washington D.C. metro area.

What you’ll do: 

• Develop and maintain computational infrastructure and predictive tools for AI biosensor optimization, including tool development for modeling fluorescence properties and biochemical performance from sequence and structure 

• Design and execute rigorous comparative experiments between model architectures.

• Collaborate with other team members to ensure seamless integration of computational and experimental aspects of the project. 

• Apply machine learning, AI techniques, and software engineering best practices to deliver scalable, maintainable, and reproducible AI systems for protein engineering.

• Carefully document data, code, and processing pipelines to enable seamless reproduction and extension of research results.

• Actively contribute to the latest advancements in the field and continuously improve your skillset with the latest advances in AI research and technologies. 

• Collaborate with interdisciplinary teams, potentially mentor junior engineers, and direct or assist in directing the work of others to meet project goals while advising stakeholders on data strategies and best practices.  

What you bring: 

• Minimum requirements: PhD in Bioengineering, Computer Science, Data Science, Statistics, Applied Mathematics, or a related field; or an equivalent combination of education and relevant experience. 

• 3+ years of experience in developing and fine-tuning deep learning models. 

• Strong programming skills in Python, PyTorch, and JAX. 

• Familiarity with protein modeling deep learning frameworks (e.g., AlphaFold, ESM, Chai-1, Boltz-1). 

• Familiarity with computer vision deep learning frameworks (e.g., SAM, cellpose).

• Experience with ML model deployment, workflow orchestration, and high-throughput data processing.