Agentic AI Engineer: Build Real-World AI Agents

SmartRecruiters, Inc.; Lisboa; Híbrido; EUR 90 000 - 150 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

28 vacation days

Private health insurance (family)

Daily meal allowance

Coverflex allowance

Flexible office hours

Resumo da oferta

SIXT is seeking an experienced Data Science professional to own the CARA agent ecosystem end-to-end, including prompts, orchestration, tools, and deployment. You will collaborate with product and customer service to translate business problems into agent capabilities and ensure robust handover into Salesforce and Genesys.

You will design for performance and cost, implement evaluation rubrics, and drive production readiness with multi-provider LLM routing and guardrails.

Qualificações

• Strong software engineering background with production ownership.
• Hands-on with Python and cloud experience.
• Experience building tool-calling agents in LangGraph/LangChain or similar.
• Familiarity with frontier/open models (Claude, OpenAI, Bedrock) and prompt engineering as an engineering practice.

Responsabilidades

• Own agents end to end: prompts, orchestration, tools, evals, deployment, dashboards.
• Translate business problems into agent capabilities with cross-functional teams.
• Build for real transactions with authentication, guardrails, and handover into Salesforce/Genesys.
• Extend evaluation harness with LLM-as-judge rubrics and A/B prompt suites.
• Balance custom agent logic with reusable platform capabilities.

Conhecimentos

Python

AWS

LangGraph/LangChain

LLMOps

Prompt engineering

Model evaluation

Ferramentas

AWS Bedrock

OpenAI

Anthropic Claude

OpenAPI

Descrição da oferta de emprego

SIXT is seeking an experienced Data Science professional to own the CARA agent ecosystem end-to-end, including prompts, orchestration, tools, and deployment. You will collaborate with product and customer service to translate business problems into agent capabilities and ensure robust handover into Salesforce and Genesys.

You will design for performance and cost, implement evaluation rubrics, and drive production readiness with multi-provider LLM routing and guardrails.