Agentic Control Center for Data Product Optimization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914384367845376 |
|---|---|
| author | Tamilselvan, Priyadarshini Bramble, Gregory Shirai, Sola Wong, Ken C. L. Chowdhury, Faisal Samulowitz, Horst |
| author_facet | Tamilselvan, Priyadarshini Bramble, Gregory Shirai, Sola Wong, Ken C. L. Chowdhury, Faisal Samulowitz, Horst |
| contents | Data products enable end users to gain greater insights about their data by providing supporting assets, such as example question-SQL pairs which can be answered using the data or views over the database tables. However, producing useful data products is challenging, and typically requires domain experts to hand-craft supporting assets. We propose a system that automates data product improvement through specialized AI agents operating in a continuous optimization loop. By surfacing questions, monitoring multi-dimensional quality metrics, and supporting human-in-the-loop controls, it transforms data into observable and refinable assets that balance automation with trust and oversight. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_10133 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Agentic Control Center for Data Product Optimization Tamilselvan, Priyadarshini Bramble, Gregory Shirai, Sola Wong, Ken C. L. Chowdhury, Faisal Samulowitz, Horst Artificial Intelligence Data products enable end users to gain greater insights about their data by providing supporting assets, such as example question-SQL pairs which can be answered using the data or views over the database tables. However, producing useful data products is challenging, and typically requires domain experts to hand-craft supporting assets. We propose a system that automates data product improvement through specialized AI agents operating in a continuous optimization loop. By surfacing questions, monitoring multi-dimensional quality metrics, and supporting human-in-the-loop controls, it transforms data into observable and refinable assets that balance automation with trust and oversight. |
| title | Agentic Control Center for Data Product Optimization |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2603.10133 |