AI Engineer - GenAI Agents for Financial Back-Office

Inetum; Lisboa; Híbrido; EUR 55 000 - 85 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

Hybrid work regime

Office in Lisbon or Porto

Professional growth opportunities

Resumo da oferta

Inetum is seeking an AI Engineer to join its team, contributing to a culture of innovation. The role focuses on building production-ready AI agents for back-office use cases in a regulated financial environment.

You will work with Python, LangGraph, LangChain, Docker and Kubernetes, design prompts, test agents, and collaborate with business analysts to translate operations like settlements and reconciliations into agent behavior, with hybrid work in Lisbon or Porto.

Qualificações

• 3–6 years of Python engineering experience.
• Experience with at least one agent framework (CrewAI, LangGraph, LangChain or AutoGen).
• Solid grasp of prompt engineering, RAG patterns and LLM evaluation techniques.
• Comfortable consuming REST/gRPC APIs and async Python.

Responsabilidades

• Design, build, and improve AI agents, including reasoning flows and tool use.
• Develop agents in Python using LangGraph, LangChain, and internal SDK.
• Design agent workflows with task decomposition, API tool calls, memory/state handling and guardrails.
• Manage prompts and configurations via Kubernetes ConfigMaps with release discipline.
• Run agent evaluation using datasets, regression suites, and quality metrics.
• Instrument agents with observability: logging, token usage, latency, and failure modes.

Conhecimentos

Python

LLM / GenAI

Agent framework knowledge

Prompt engineering

Testing culture

Ferramentas

LangGraph

LangChain

CrewAI

AutoGen

Docker

Kubernetes

pytest

Descrição da oferta de emprego

Inetum is seeking an AI Engineer to join its team, contributing to a culture of innovation. The role focuses on building production-ready AI agents for back-office use cases in a regulated financial environment.

You will work with Python, LangGraph, LangChain, Docker and Kubernetes, design prompts, test agents, and collaborate with business analysts to translate operations like settlements and reconciliations into agent behavior, with hybrid work in Lisbon or Porto.