AI Engineer

QUANTEAM - Portugal (RAINBOW PARTNERS Group); Lisboa; Híbrido; EUR 15 000 - 39 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

Health Insurance

Extra holiday: Christmas Eve or New Y'

Access to Quanteam Academy

Elevate Program

Hybrid work model (Lisbon/Porto)

Resumo da oferta

Quanteam, founding entity of Rainbow Partners, is seeking an AI Agent Engineer to design, build and improve AI agents for Custody Back-Office use cases. You will work on agent reasoning, workflows, tool integration, prompts and evaluation.

The role requires 3-6 years of Python engineering with 1-2 years in AI applications, and experience with LangGraph/LangChain/CrewAI/AutoGen. Hybrid work in Lisbon or Porto (2 days/week) and a gross salary of 15,000–38,500 EUR per year, with a permanent

Qualificações

• 3-6 years of professional experience in Python engineering, including at least 1-2 years working with LLM or Generative AI applications.
• Hands-on experience with at least one AI agent framework such as LangGraph, LangChain, CrewAI, AutoGen or similar.
• Strong understanding of AI agent concepts, including ReAct-style reasoning, tool/function calling, structured outputs, memory and state management.
• Solid knowledge of Prompt Engineering, RAG architectures and LLM evaluation techniques.
• Experience integrating and consuming REST or gRPC APIs, working with JSON schemas and asynchronous Python.
• Working knowledge of Docker and Kubernetes, including deployment basics, log analysis and configuration management through ConfigMaps.
• Strong software engineering and testing practices, including pytest, mocking LLM calls and designing deterministic tests for non-deterministic systems.
• Availability to work in a hybrid model in Lisbon or Porto (2 days/week).
• This role offers a gross annual salary between 15.000€ to 38.500€,

Responsabilidades

• Design, develop and maintain AI agents in Python using LangGraph, LangChain and internal SDKs.
• Design agent workflows, including task decomposition, reasoning flows, tool/function calling, memory and state management, guardrails and human-in-the-loop checkpoints.
• Adapt central general-purpose agents to local Custody Back-Office use cases, considering business workflows, formats, vocabulary and operational controls.
• Engineer, test and version prompts and configurations, ensuring proper release management through Kubernetes ConfigMaps.
• Build and maintain agent evaluation frameworks, including golden datasets, regression testing, quality metrics (accuracy, groundedness and escalation rate) and shadow testing before production rollout.
• Implement observability and monitoring capabilities, including structured logging, reasoning traces, token consumption, latency and failure mode analysis.
• Work closely with Business Analysts and technical teams to translate operational knowledge related to settlements, corporate actions, client queries and reconciliations into effective AI agent behaviour.

Conhecimentos

Python engineering

LLM / Generative AI

LangGraph

LangChain

REST APIs

gRPC

Docker

Kubernetes

JSON schemas

Pytest

Ferramentas

LangGraph

LangChain

CrewAI

AutoGen

Descrição da oferta de emprego

Quanteam, founding entity of RAINBOW PARTNERS, is a consulting firm specialized in Banking and Financial Services. With more than 1200 consultants across 10 countries, we combine business and IT expertise to support leading institutions in transformation projects.

Our core areas:

Operations & Finance

We are looking for an AI Agent Engineer to design, build and continuously improve AI agents for Custody Back-Office use cases.

You will work on agent reasoning, workflows, tool integration, prompts and evaluation, adapting central general-purpose AI agents to specific operational and business requirements.

Main Responsibilities

• Design, develop and maintain AI agents in Python using LangGraph, LangChain and internal SDKs.
• Design agent workflows, including task decomposition, reasoning flows, tool/function calling, memory and state management, guardrails and human-in-the-loop checkpoints.
• Adapt central general-purpose agents to local Custody Back-Office use cases, considering business workflows, formats, vocabulary and operational controls.
• Engineer, test and version prompts and configurations, ensuring proper release management through Kubernetes ConfigMaps.
• Build and maintain agent evaluation frameworks, including golden datasets, regression testing, quality metrics (accuracy, groundedness and escalation rate) and shadow testing before production rollout.
• Implement observability and monitoring capabilities, including structured logging, reasoning traces, token consumption, latency and failure mode analysis.
• Work closely with Business Analysts and technical teams to translate operational knowledge related to settlements, corporate actions, client queries and reconciliations into effective AI agent behaviour.

Technical Requirements

• 3-6 years of professional experience in Python engineering, including at least 1-2 years working with LLM or Generative AI applications.
• Hands-on experience with at least one AI agent framework such as LangGraph, LangChain, CrewAI, AutoGen or similar.
• Strong understanding of AI agent concepts, including ReAct-style reasoning, tool/function calling, structured outputs, memory and state management.
• Solid knowledge of Prompt Engineering, RAG architectures and LLM evaluation techniques.
• Experience integrating and consuming REST or gRPC APIs, working with JSON schemas and asynchronous Python.
• Working knowledge of Docker and Kubernetes, including deployment basics, log analysis and configuration management through ConfigMaps.
• Strong software engineering and testing practices, including pytest, mocking LLM calls and designing deterministic tests for non-deterministic systems.
• Availability to work in a hybrid model in Lisbon or Porto (2 days/week).
• This role offers a gross annual salary between 15.000€ to 38.500€, defined according to the candidate's experience and competencies.
• Permanent contract (CTI).
• Health Insurance.
• Extra holiday: Christmas Eve or New Year's Eve.
• Access to Quanteam Academy for continuous learning and certifications.
• Elevate Program to support professional development and career growth.
• Ongoing follow-up and technical support provided by a Consultant Expert.

Hybrid position based in Lisbon and Porto