Junior Software Developer

University of Chicago•Chicago, IL

5d•Hybrid

About The Position

The Junior Software Developer will join the Globus Labs research and software development team at the University of Chicago, contributing to a new effort applying AI interpretability to scientific discovery. The role combines mechanistic interpretability, scientific foundation models, and open data infrastructure, and includes research and engineering work to help reveal scientific insights in areas such as materials science, chemistry, catalysis, and x-ray science. The position is wholly or partially supported by contractual grant funding; continued employment depends on receipt of that funding and satisfactory performance. Under the direction of the Principal Investigator, the developer will help coordinate the effort’s daily work, contribute to its research roadmap, and serve as a hands-on contributor across research and engineering projects. The role includes implementing interpretability methods, building reusable tools and GPU-enabled workflows, supporting foundation model and scientific automation initiatives, and developing and maintaining cloud-based software and data services. Remote work is available, with occasional in-person meetings required.

Requirements

College or university degree in a related field.
Knowledge and skills developed through less than two years of work experience in a related job discipline.
Excellent written and verbal communication skills in English.
Ability to work effectively in a dynamic, hybrid work environment and collaborate with team members.

Nice To Haves

Research internships, research assistantships, or co-op positions in machine learning, AI interpretability, computational science, or software engineering.
Internship experience at a national laboratory, academic research group, or industry AI lab.
Hands-on research project experience with large language models, scientific foundation models, or other deep learning systems, including thesis work, conference papers, or preprints.
Contributions to open-source projects, particularly in machine learning, interpretability tooling, or scientific computing.
Experience developing, testing, and deploying software in a collaborative team setting.
Exposure to research outputs such as co-authored papers, workshop submissions, technical reports, posters, or conference presentations.
Python development experience, with knowledge of Node.js and/or a compiled language such as C/C++.
Experience with deep learning frameworks such as PyTorch or JAX and familiarity with model libraries such as Hugging Face Transformers.
Familiarity with mechanistic interpretability methods such as sparse autoencoders or dictionary learning, probing, circuit analysis, activation patching, or activation steering.
Experience with TransformerLens, nnsight, or SAELens.
Understanding of transformer architectures and how large language models and scientific foundation models are trained and evaluated.
Experience running GPU-accelerated workloads on HPC clusters or cloud GPU instances and familiarity with job schedulers such as Slurm.
Knowledge of cloud computing platforms such as AWS, GCP, or Azure.
Proficiency in software development practices, including Git, code review, testing, documentation, and CI/CD pipelines.
Experience building or consuming REST APIs and command-line tools; familiarity with asynchronous or socket-based interfaces.
Experience with NumPy, pandas, scikit-learn, Jupyter, and handling large datasets.
Interest in or exposure to scientific domains such as materials science, chemistry, catalysis, or x-ray science.
Strong analytical and problem-solving skills.
Interest in and experience developing software to support data-driven research.

Responsibilities

Coordinate day-to-day activities for the interpretability-for-scientific-discovery effort under the Principal Investigator’s direction.
Track workstreams, milestones, and deliverables, and raise issues to senior staff when needed.
Assist in developing the group’s research roadmap and priorities in alignment with broader laboratory goals.
Contribute hands-on to research and engineering across group projects and help maintain consistent technical quality.
Develop tutorials and onboarding materials, and organize group check-ins and technical discussions.
Implement and apply mechanistic interpretability techniques to scientific foundation models and AI systems, including feature discovery, sparse dictionary learning, circuit analysis, probing, and activation steering.
Investigate how models trained on scientific data represent physical, chemical, and materials concepts, and help evaluate the reliability of interpretability methods in these settings.
Build and maintain reusable tooling, pipelines, and GPU-accelerated workflows for interpretability experiments.
Support foundation model training, agentic oversight frameworks, and autonomous laboratory efforts.
Contribute to laboratory initiatives and collaborations, including Globus GPU-accelerated inference, ISAAC and other Genesis programs, Terminal-Bench Science, and new laboratory and industry collaborations.
Contribute to developing, deploying, and managing data infrastructure services, including systems for publishing large datasets.
Help define and document requirements and architecture for secure, scalable, asynchronous data access services.
Develop, test, document, and maintain software on a cloud-based platform, including REST, command-line, and socket-based interfaces.
Apply GitOps principles for reproducible deployments.
Integrate cross-service features with other development teams and troubleshoot day-to-day technical problems using standard procedures.
Support external engagement, including partnership discussions, technical meetings, and relationships with program managers and collaborators.
Identify real-world applications for the group’s technologies and contribute ideas for new projects or exploratory efforts.
Test code components and help ensure implementation standards are met.
Perform other related work as needed.