Serhii Zautkin
Serhii Zautkin is a Programmer Analyst at the Canadian Astronomy Data Centre (CADC / NRC). He specializes in full-stack platform development and modernizing astronomy software, with a focus on blending AI capabilities directly into science platforms. He holds a B.Sc. from the University of Manitoba.
Session
Two options exist today, and each falls short in a different direction. Point a commercial AI application at data held on a science platform and the science turns into plumbing: credentials, staging, transfers, all for a dataset that should never have moved. Run a coding CLI inside the platform instead and it sits beside the data but sees only a shell and a filesystem — nothing of the archive, the session lifecycle, or the access rules governing them. The first cannot reach the data; the second cannot see the platform. Neither was built for research.
The scientific workbenches now appearing run on the researcher's own machine or cluster so the data need not move. But their domain readiness is built for the life sciences, and the integration path is proprietary: the platform arrives as a plug-in inside the vendor's application, and research context accumulates in that vendor's formats. Astronomy already has platforms tuned over decades to its data models, access rules and standards; we should not have to migrate onto a general-purpose workbench to benefit from these models — the models should meet our platforms on our terms.
I illustrate this with a client over CANFAR, Skaha and the CADC archives whose agent tools are not a parallel API: each performs the same action as a control in the interface, so astronomers and agents share one surface. Both launch, monitor and stop interactive and headless sessions, and the agent can task one with a script in an image carrying the astronomy libraries. Retrieved observations collected in a research area, products land in VOSpace, FITS and cube viewers give a fast visual read the model can be asked to assess, and workflows serve as written protocols the agent follows.
As this work is ongoing, the exact final vision for the user-facing product is still actively taking shape. Rather than presenting a finished system, we present our integration across CANFAR, Skaha, and CADC as the foundation and core of a future AI-enabled science platform. We bring these developments to ADASS to open a dialogue with the community—seeking to learn how researchers envision the natural blend of AI models and research workflows, and how our shared platforms can best deliver that vision.