Data & AI Agent Architect – GCP / Python (M/F)

Project overview

Cristal Solutions is looking for a Data & AI Agent Architect for a mission within a Data platform on GCP.

You will design and develop AI agents for technical teams to facilitate understanding of pipelines, automate certain development tasks, and support production deployments.

Your missions will include integrating agents with APIs and Data components, developing reusable tools, and improving development practices. You will implement the necessary tests, evaluations, and monitoring to ensure the reliability of solutions and control the actions performed by the agents.

You will also contribute to the modernization of the technical foundation, the security of deployments, and the improvement of observability, in collaboration with the Data and platform teams.

Mission based in IDF, with three days per week on-site. Start as soon as possible.

Profile Sought

Your Missions

Essential Skills

Appreciated Skills

Do you have a profile as an AI Engineer, Data Engineer, Software Engineer, or Architect, with experience in coding and AI agents? Send us your CV, specifying your availability, your daily rate, and an example of an agent delivered in production.

Work Environment

Environment: GCP, Python, Java/Scala, BigQuery, dbt, Kafka Streams, Dataproc, APIs, and CI/CD. Mastery of all these technologies is not required.

• Design, develop, and deploy AI agents integrated into the platform.
• Connect agents to tools, APIs, and Data components.
• Develop reusable components and improve developer tooling.
• Implement the tests, monitoring, and controls necessary for reliable production use.
• Contribute to the modernization, security, and observability of the technical foundation.
• Solid experience in software development or platform engineering on GCP.
• Mastery of industrialization practices: CI/CD, testing, and observability.
• Experience with batch processing using dbt/BigQuery or streaming using Kafka.
• Personally designed and delivered at least one AI agent in production, used by others.
• Development of MCP servers.
• Use of agent frameworks: Google ADK, Claude Agent SDK, LangGraph, or equivalents.
• Experience with internal tools used by multiple teams.
• Implementation of agent or LLM system evaluations.

Project details

• Start: ASAP
• Duration: 24 mois renouvelable
• Remote: hybrid
• Location: Île-de-France, France, FR
• Languages: French