Frontier Agentic AI Engineer - Production-Grade Multi-Agent Systems

santander; Santa Bàrbara; Presencial; EUR 90.000 - 140.000; Jornada completa

Descripción de la vacante

Santander AI Lab is seeking an experienced Frontier Agentic AI Engineer based in Madrid, Spain. You will design and deploy production-grade, memory-enabled multi-agent AI systems, and refine LLMs for banking use cases, coordinating with researchers and business stakeholders to generate tangible bank value.

You will prototype in two-week sprints and integrate new tools into the lab workflow, ensuring end-to-end code quality and readiness for production scaling.

Formación

• 4-8 years of software engineering or AI engineering experience with 2+ years building and maintaining LLM-powered systems in production.

Responsabilidades

• Design and deploy production-grade agentic AI systems with memory, planning, tool use and error recovery.
• Develop and fine-tune small and medium language models (SLMs) for regulated banking use cases.
• Architect MCP servers, A2A protocols, and federated API layers for multi-country infrastructure.
• Prototype ideas in two-week sprints with architecture, code and demos.
• Collaborate with researchers, data scientists and stakeholders to translate tech concepts into bank value.
• Monitor emerging research and integrate new tools into the lab workflow.
• Produce clean, tested, observable code and hand off to AI Science for production scaling.

Conocimientos

Multi-agent systems

REST APIs

Production-grade code

Memory management

AI engineering

Educación

Bachelor's or Master's in CS/SE

Advanced AI qualification (Masters)

Herramientas

LoRA

SFT

RLHF

Descripción del empleo

Santander AI Lab is seeking an experienced Frontier Agentic AI Engineer based in Madrid, Spain. You will design and deploy production-grade, memory-enabled multi-agent AI systems, and refine LLMs for banking use cases, coordinating with researchers and business stakeholders to generate tangible bank value.

You will prototype in two-week sprints and integrate new tools into the lab workflow, ensuring end-to-end code quality and readiness for production scaling.