Skip to content
Job boardOpen

Agentic Foundation Governance Expert

Original title: Expert gouvernance socle agentique

Unitech Solutions · Saint-Denis, France · Hybrid · SENIOR

  • Saint-Denis, France · Hybrid
  • French

Indexed (today) · refreshed constantly

Budget
€75 / h
Start
flexible
Work mode
Hybrid
Contract
Freelance / B2B
Seniority
SENIOR

Required skills

Nice to have:IPAMCMDBMonitoring ToolsTicketing ToolsBGPOSPFVRF

Languages: French

Job description

Seeking an expert in agentic foundation governance to design and implement the agentic architecture for the AI Network Architecture Copilot. The role involves setting up a principal agent to orchestrate specialized agents and defining governance rules for their autonomy levels.

Project overview

Context

Intervention within the IA Prod Directorate, as part of the agentic foundation, for the Architecture Department of our client.

Missions

Working Conditions

Expected Deliverables

Technologies / skills & soft skills

Technical Environment

Soft skills

  • Design and implementation of the agentic architecture for the AI Network Architecture Copilot
  • Setup of a principal agent capable of orchestrating several specialized agents: Architecture Agent, Network Knowledge Agent, Network Troubleshooting Agent, Automation Agent, and Security Agent
  • Proposal of the target technical architecture and the agent orchestration foundation
  • Definition of the autonomy level model, from level 1 documentary assistant to level 5 autonomous agent
  • Definition of governance rules associated with the different autonomy levels
  • Definition of performance indicators and establishment of the baseline before deployment
  • Measurement of DAT production time, incident analysis time, documentary search time, and adoption rate
  • Contribution to the implementation roadmap, resource and workload estimation, as well as operational and evolution modalities
  • Development of the principal agent and the orchestration, planning, and tool calling mechanisms
  • Design of read-only connectors to NetBox, Git, IPAM, CMDB, monitoring tools, and ticketing tools
  • Setup of the AI Network Lab test environment for experimentation without impacting production
  • Integration of the POC conversational interface
  • Generation of architecture deliverables from exchanges and available knowledge
  • Definition and implementation of security controls: RBAC authorizations, perimeter separation, protection of configurations and sensitive information
  • Implementation of human validation mechanisms based on the principle: AI proposes, human validates, AI executes, AI controls and traces the result
  • Implementation of traceability for consultations, reasoning, and agent actions
  • Setup of supervision and error management
  • Construction of evaluation scenarios and agent safeguards
  • Measurement of POC success criteria
  • Documentation of technical choices, architectures, governance rules, and security mechanisms
  • Transfer of practices to network architects, AI teams, and security teams
  • Integration of work into an industrializable trajectory
  • Collaboration with DRC-DA architects and referents, the IA Prod Directorate, the IA Product Owner, the security referent, and pilot architects
  • No actions or modifications performed on production network infrastructures during the mission
  • Connections to tools performed exclusively in read-only mode
  • All future actions subject to prior human validation
  • 3 months renewable
  • Location: 93210 Saint-Denis
  • Max daily rate: €600
  • Number of positions: 2
  • Experience: > 7 years
  • Target technical architecture document
  • Agent orchestration foundation proposal
  • Autonomy level model
  • Agentic governance rules
  • Security constraints analysis
  • Principal agent for the POC Architecture Copilot
  • First specialized agents for the POC
  • Read-only connectors to NetBox, Git, IPAM, CMDB, monitoring, and ticketing tools
  • AI Network Lab test environment
  • POC conversational interface
  • Human validation mechanisms
  • Traceability and supervision mechanisms
  • Agent evaluation scenarios and safeguards
  • POC evaluation report
  • KPIs and baseline before deployment
  • Implementation roadmap
  • Resource and workload estimation
  • Proposal for operational and evolution modalities
  • Updated tracking table
  • Supports and minutes from steering committee meetings
  • Documentation of technical choices and implemented practices
  • Agentic engineering
  • Multi-agent orchestration
  • LangGraph or equivalent framework
  • Generative Artificial Intelligence
  • LLM
  • Prompting
  • Tool calling
  • MCP protocol
  • Python
  • API development
  • REST API
  • NetBox
  • Git
  • IPAM
  • CMDB
  • Monitoring tools
  • Ticketing tools
  • RBAC
  • Authorization management
  • Traceability and observability of AI agents
  • NetDevOps
  • IP networks and network automation
  • BGP
  • OSPF
  • VRF
  • Kubernetes
  • Application deployment
  • Critical network and information system architectures
  • Zero Trust Architectures
  • Governance and compliance of AI solutions
  • Rigor and sense of service
  • Excellent analytical and synthesis skills
  • Autonomy and ability to take charge of complex subjects
  • Interest in production AI topics and agentic engineering
  • Proactive and taste for continuous improvement
  • Interpersonal skills with network architects, AI teams, and security teams
  • Ability to collaborate with multiple technical and business stakeholders
  • Pedagogical skills and ability to transfer practices
  • Strong sensitivity to security, governance, and compliance issues
  • Ability to evolve in a complex and innovative environment
  • Ability to formalize decisions and produce clear documentation
  • Applied network knowledge

Project details

  • Start: flexible
  • Remote: hybrid
  • Location: Saint-Denis, France, FR
  • Rate: 75-75 EUR/h
  • Languages: French

Indexed by 9am's meta-search. You are a job provider? Claim or manage this listing

Found on collective.work

  1. Job boardNew

    GenAI Architect

    Xebia· Wrocław, Sucha 3, Poland· Remote· Senior· zł 29–34 / h

    • Kubernetes
    • Model Context Protocol
    • LangGraph
    • +9

Browse all jobs
Sign up free to apply