Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.12483 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917337761841152 |
|---|---|
| author | Maddi, Aadyaa Naval, Prakhar Mande, Deepti Duan, Shane Girish, Muckai Sekar, Vyas |
| author_facet | Maddi, Aadyaa Naval, Prakhar Mande, Deepti Duan, Shane Girish, Muckai Sekar, Vyas |
| contents | Across many domains (e.g., IoT, observability, telecommunications, cybersecurity), there is an emerging adoption of conversational data analysis agents that enable users to "talk to your data" to extract insights. Such data analysis agents operate on timeseries data models; e.g., measurements from sensors or events monitoring user clicks and actions in product analytics. We evaluate 6 popular data analysis agents (both open-source and proprietary) on domain-specific data and query types, and find that they fail on stateful and incident-specific queries. We observe two key expressivity gaps in existing evals: domain-customized datasets and domain-specific query types. To enable practitioners in such domains to generate customized and expressive evals for such timeseries data agents, we present AgentFuel. AgentFuel helps domain experts quickly create customized evals to perform end-to-end functional tests. We show that AgentFuel's benchmarks expose key directions for improvement in existing data agent frameworks. We also present anecdotal evidence that using AgentFuel can improve agent performance (e.g., with GEPA). AgentFuel benchmarks are available at https://huggingface.co/datasets/RockfishData/TimeSeriesAgentEvals. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12483 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Generating Expressive and Customizable Evals for Timeseries Data Analysis Agents with AgentFuel Maddi, Aadyaa Naval, Prakhar Mande, Deepti Duan, Shane Girish, Muckai Sekar, Vyas Artificial Intelligence Machine Learning Across many domains (e.g., IoT, observability, telecommunications, cybersecurity), there is an emerging adoption of conversational data analysis agents that enable users to "talk to your data" to extract insights. Such data analysis agents operate on timeseries data models; e.g., measurements from sensors or events monitoring user clicks and actions in product analytics. We evaluate 6 popular data analysis agents (both open-source and proprietary) on domain-specific data and query types, and find that they fail on stateful and incident-specific queries. We observe two key expressivity gaps in existing evals: domain-customized datasets and domain-specific query types. To enable practitioners in such domains to generate customized and expressive evals for such timeseries data agents, we present AgentFuel. AgentFuel helps domain experts quickly create customized evals to perform end-to-end functional tests. We show that AgentFuel's benchmarks expose key directions for improvement in existing data agent frameworks. We also present anecdotal evidence that using AgentFuel can improve agent performance (e.g., with GEPA). AgentFuel benchmarks are available at https://huggingface.co/datasets/RockfishData/TimeSeriesAgentEvals. |
| title | Generating Expressive and Customizable Evals for Timeseries Data Analysis Agents with AgentFuel |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2603.12483 |