Data Scientist

Caltech is a world-renowned science and engineering institute that marshals some of the world's brightest minds and most innovative tools to address fundamental scientific questions. We thrive on finding and cultivating talented people who are passionate about what they do. Join us and be a part of the diverse Caltech community.

Job Summary

NExScI at Caltech/IPAC is seeking a data scientist to join our software development team and drive the continued development of AstroFetch, an AI-powered assistant that helps astronomers query, analyze, and visualize data from astronomical archives. This is a technical software position focused on developing, testing, and maintaining the AI tooling, data interfaces, and analysis tools behind AstroFetch. It is an excellent fit for a developer or data scientist who enjoys writing high-quality code, wants to build applications with large language models, and has enough familiarity with astronomy or another physical science to create tools that researchers can trust.

AstroFetch uses large language models to translate natural-language questions into archive queries, data analysis, and visualizations. A public beta version already exists, built in Python and supporting multiple LLM providers (Anthropic Claude, OpenAI, Google Gemini, and local models). You will develop the software that makes its answers accurate and useful: building the prompts, tools, and retrieval systems it relies on, creating automated evaluations that measure its performance, and extending it to new data sources and capabilities. You will work closely with software and infrastructure engineers, who share responsibility for deploying, securing, and operating the service, and with NExScI scientists, who provide domain expertise and user feedback.

This role will work within a vibrant team of scientists and developers at the NASA Exoplanet Science Institute (NExScI). As a part of IPAC, NExScI (nexsci.caltech.edu) provides archive services, community support, science operations, and analysis tools related to the discovery and characterization of planets beyond our solar system (exoplanets) using data from observatories in space and on the ground. IPAC also hosts several other NASA and Caltech data archives, and this position will have the opportunity to bring AI-driven tools and approaches to those projects as well.

Essential Job Duties

Your primary focus will be the scientific quality and capabilities of AstroFetch, while also exploring new applications of AI across IPAC. Key responsibilities:

• Improve how AstroFetch answers scientific questions through prompt design, tool design, and retrieval-augmented generation (RAG) over archive documentation and data schemas.
• Build evaluation datasets and benchmarks that measure the scientific accuracy of AstroFetch’s queries, analyses, and answers, and use them to guide development and compare LLM providers.
• Develop and validate the analysis and visualization tools the assistant uses with catalogs, time series, spectra, and other archive data products, including the move from static plots to interactive visualizations (e.g., Bokeh or Plotly).
• Connect AstroFetch to additional archives (such as the Keck Observatory Archive, and IRSA), working with the scientists and archive teams who know those data best.
• Write clean, tested, well-documented Python and contribute to a shared codebase through version control, code review, and continuous integration, in partnership with the engineers who lead deployment, security, and infrastructure.
• Gather feedback from astronomers using AstroFetch, analyze how it is used, and turn what you learn into improvements.
• Identify, prototype, and evaluate new applications of AI and large language models across IPAC’s data archives and scientific workflows.
• Stay current with developments in AI for science, share them with colleagues, and present AstroFetch and related work to the astronomy community, including at conferences and in publications where appropriate.

Basic Qualifications

If you have the following in your background, then we want to hear about your interest in joining our team:

• Bachelor’s degree in Computer Science, Data Science, Astronomy, Physics, or a related technical field.
• Strong Python skills, including the scientific Python ecosystem (e.g., NumPy, pandas, Astropy, Matplotlib).
• Experience working with astronomical or other large scientific datasets, including querying data with SQL or a similar language (e.g., ADQL).
• Hands-on experience using large language model APIs or AI tools in code (e.g., Anthropic Claude, OpenAI, Google Gemini); experience from research or personal projects is welcome.
• Strong communication skills and enthusiasm for working with both scientists and software engineers.

Preferred Qualifications

These additional qualifications may give you a strong start, though we still encourage you to apply even if you don’t have all of them:

• Master’s degree in Computer Science, Data Science, Astronomy, Physics, or a related technical field.
• Research experience in exoplanets, time-domain astronomy, or observational astronomy.
• Familiarity with astronomical archives and Virtual Observatory standards (TAP, ADQL) and tools such as astroquery or PyVO.
• Experience building retrieval-augmented generation (RAG) or agentic AI applications with tool use and structured outputs.
• Experience designing evaluations or benchmarks for machine learning or LLM systems.
• Experience with interactive data visualization libraries (Bokeh, Plotly, or similar).
• Experience with Python web frameworks (FastAPI, Flask), Git-based collaboration, automated testing, and CI.
• Familiarity with Docker, PostgreSQL, and cloud services.
• A record of open-source scientific software, public code, or publications.

Required Documents

• Full resume or CV
• Cover letter
• Optional: Provide links to code (e.g., GitHub) or relevant publications

Application Details

• Hybrid position: Typically, three to four days per week on the Caltech campus in Pasadena, California, and one to two days remote, depending on operational needs. Depending on start date, an initial remote working arrangement may be considered.
• Applications will be reviewed on a rolling basis until the position is filled.

Life at IPAC and Caltech

Caltech is an award-winning workplace and one of the best places to work in the greater Los Angeles area. We offer an exciting and dynamic work environment, career development opportunities, competitive pay programs, great benefits, and the opportunity to participate in many campus programs and activities.

People choose to work at IPAC for many reasons, and the casual, employee-centric culture often leads to fulfilling, long-term careers and lasting relationships. We seek the best and brightest talent to join our workforce and become members of our community.