Data exploration
Analyze real-world, multi-file datasets and open-ended questions with DS-Star, or plan a biomedical analysis before coding it with DSWizard — both produce full research-style reports instead of only short answers.
Work across different file formats and combine information from multiple sources
Handle messy, real-world datasets and open-ended questions, not just clean single-table queries
Produce full research-style reports instead of only short answers
Plan an analysis before coding it, for cases where reliability matters more than speed
Data exploration agents
Our Co-Scientist acts as a collaborative partner designed to work side by side with computational scientists in healthcare, engineering, or product. It offers fast, interpretable, and actionable insights from your and public data. It is named after the Border Collie which is known to herd with intelligence, agility, and responsiveness.
Data-science runtime optimized for iterative coding and streamed analysis in the co-scientist loop.
DS-Star is a data-science agent designed for real-world analysis tasks that involve messy, multi-file datasets and open-ended questions. It can work across different file formats, combine information from multiple sources, and produce full research-style reports instead of only short answers. In evaluations, DS-Star performs especially well on harder tasks that require multi-file reasoning and code-based analysis.
Nam et al. (2025). DS-STAR: Data Science Agent for Solving Diverse Tasks across Heterogeneous Formats and Open-Ended Queries. arXiv preprint.
Two-phase biomedical data science agent (planning then implementation).
DSWizard is a biomedical data-science agent designed to improve reliability by planning before coding. Instead of jumping straight to code generation, it first drafts and refines an analysis plan with the user, then implements the plan in code. This workflow helps reduce errors in AI-generated analyses and visualizations and supports researchers in collaboratively building accurate analysis pipelines.
Wang et al. (2026). Making large language models reliable data science programming copilots for biomedical research. Nature Biomedical Engineering.
Question
🚢 Using the /app/custom_data/titanic.csv dataset, write a short report for a general audience with: 1) overall survival rate, 2) survival rate by sex, 3) survival rate by passenger class (Pclass)
Expected answer
Overall survival: 38.38% (342/891). By sex: female 74.20% (233/314), male 18.89% (109/577). By class: 1st 62.96% (136/216), 2nd 47.28% (87/184), 3rd 24.24% (119/491).
No. These are agents that reason over your files and write code to analyze them, producing a written report rather than a fixed dashboard.
Messy, multi-file, multi-format datasets and open-ended questions — the kind of real-world data exploration that does not fit a single clean spreadsheet.
Data is encrypted at rest. Attachments you upload into the agent workspace are deleted after a period of time, and we will let you know when that happens. We recommend not uploading personally identifiable information or sensitive data.
Have a question? Please get in touch.