Speaker
Description
We propose an on-premises, agentic system built on open-weight models to automate repetitive tasks in grid-based analyses and MC production, cutting manual effort for submitting jobs for both operators and experts. The agent integrates with existing middleware (in the case of ALICE: Hyperloop, MonALISA, and jAliEn), using retrieval-augmented generation to interpret production and analyses requests and policies. A planner-executor loop coordinates end-to-end tasks of input preparation, job submission, monitoring, adaptive retry/backoff for transient site issues, output collection and artifact registration, while maintaining durable state (job registry, artifact tracker) and full provenance. Safety and governance are built in via role-based access, quotas, allowlists, dry-run/approval modes, change windows and clear escalation paths. Structured event logs supporting dashboards, alerts and post-mortems will further enhance the experience of the analysers and users who will be able to understand if and why issues arose during the submission of their analysis or MC workflows.
For the collaboration, this reduces operational load, shortens time from request to validated datasets and improves reproducibility. The agent enforces policy consistently (e.g. priorities, resource quotas), learns common failure modes to choose between resubmission, rerouting, or escalation and avoids orphaned or inconsistent productions.
Within the agentic-AI landscape, this system embodies a stateful tool-using planner rather than a one-shot assistant or brittle script. Emphasis is placed on reliable adapters and explicit safety rails, not model “cleverness.” Implementing open-weight models on our hardware ensures data residency, predictable cost and the ability to fine-tune on collaboration-specific schemas and log patterns - no external calls at inference time.
CERN group/ Experiment
EP-AIP
| Working area | Area 6: Large Language Models-based assistants |
|---|---|
| Project goals | Autonomous agent for analysis and MC grid management |
| Timeline | H1: Wrap grid ops as idempotent tools (submit/status/cancel/logs/register). Stand up RAG on available docs and logs. M12: MVP agent (planner→act→observe), supervised operation for analysis. M24: Scale-out and adapt for complex MC tasks, MC operation supervised by human operator. Year 3: trustworthy autonomy & handover |
| Available person power | 0.25 ORIGIN, 0.1 STAFF |
| Additional person power request | 1 ORIGIN |
| Is this an already ongoing activity? | No |
| Indicative hardware resources needs | High VRAM server for LLM (7-13B parameters) + large context, ideally of the order of 80GB of VRAM |