Frontier Agentic AI Engineer — Santander AI Lab
Core
Designing, building, and deploying production-grade agentic AI systems (multi-agent orchestration, memory, planning, tool use) and fine-tuning small language models for regulated banking use cases.
Role type
Senior IC Agentic AI Engineer
Builds
Production-grade agentic AI systems, multi-agent orchestration workflows, and fine-tuned domain-specific language models for a global bank.
Domain
Financial Services / Agentic AI / LLMs
Deliverable
production ML models
Required skills
Multi-agent system design and deployment, LLM fine-tuning (SFT, LoRA, RLHF), Python (advanced), Cloud infrastructure (AWS), API development (FastAPI), Model evaluation and observability, MCP server design, DevOps (Docker, CI/CD)
Preferred skills
Harness Engineering methodologies, Kubernetes, Terraform, Knowledge graphs/GraphRAG, A2A protocols, x402, AP2
Technologies
LangGraph, AutoGen, CrewAI, Semantic Kernel, Anthropic Claude API, OpenAI, Llama, Mistral, vLLM, Ollama, Unsloth, Harness Engineering, Kiro, Windsurf, Devin, RAGAS, Promptfoo, Langfuse
Responsibilities
Designing and deploying production-grade agentic AI systems with real memory and error recovery; Developing and fine-tuning SLMs for regulated banking use cases; Architecting MCP servers and federated API layers; Prototyping new ideas in two-week sprints; Collaborating with researchers and stakeholders to translate technical concepts into bank value; Monitoring emerging research and integrating new tools into the lab workflow.
Seniority
Senior, hands-on IC