Agentic systems and RAG
Page reviewed on
Design, production roll-out, logging and traceability of multi-agent systems.
We design and roll out systems in which several AI agents share a task, with their scope rules, their evaluation suite and their logs.
What we do
- Architecture of the agents and their orchestration
- System instruction files and scope rules, versioned in the repository
- Behaviour evaluation suite, written before go-live
- Logging of agent actions and traceability of answers
- Human validation points built into the architecture
Formats
- Scoping and architecture — 5 days.
- Prototype in real conditions — 15 to 25 days.
- Production roll-out, on quotation.
Deliverables
Architecture, instruction files, scope rules, evaluation suite, session logs.
Behaviour validation
Behaviour specification and evaluation methods are described on the page Compliance and reliability of agentic AI systems. Knowledge bases queried by agents are covered on the page Knowledge bases and RAG.
Who does the work
Fabrice Trollet, managing director of AGGIL, EFREI engineer. Background, publications and sources: References and sources.
Tax credit
AGGIL is approved for the French research tax credit (CIR) and innovation tax credit (CII) (2024-2026). For research or innovation operations entrusted to it only, expenses incurred with an approved body may be included in the client company's tax credit base, under the conditions and limits set out in Article 244 quater B of the French General Tax Code. The research and innovation tax credit offer