AI Engineer

Capitole; Barcelona; Híbrido; EUR 45.000 - 75.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Training budget €1200/year and ongoing

Private health insurance

Hybrid work model (2 days onsite)

Wellhub: fitness, wellness and mental‑

Football and paddle tennis teams

Team buildings and global events

Descripción de la vacante

Capitole in Barcelona/Spain seeks an AI Engineer to design, build, and deploy LLM-powered applications and multi-agent systems on AWS. You’ll work at the intersection of software, AI engineering, data pipelines, and cloud deployment, turning experiments into production.

You will collaborate with Data Scientists, Platform and DevOps, Data Engineers, and Product Managers. 3–5 years in software engineering with 1–2 years AI/ML, strong Python, and experience with LangChain, LangGraph, HuggingFace,

Formación

• 3–5 years of software engineering experience.
• 1–2 years focused on AI/ML engineering.
• Strong proficiency in Python.
• Experience with LLM frameworks and AI/ML tools.
• Hands-on experience with RAG systems, embeddings, vector stores, and semantic search.
• Experience with LLM APIs on AWS Bedrock, OpenAI, or Anthropic.
• Knowledge of prompt engineering, model evaluation, and CI/CD for ML.
• Familiarity with Docker, Kubernetes, and data pipelines (Airflow/Prefect).
• Experience in Scrum Agile environments and collaboration with cross-functional teams.
• Spanish language skills are a plus.

Responsabilidades

• Design, build, and maintain LLM-powered applications and multi-agent systems using LangGraph and LangChain.
• Develop and optimise RAG pipelines including ingestion, embeddings, retrieval, and vector search.
• Manage vector databases on Aurora, OpenSearch, Pinecone, or equivalents.
• Build data/ETL pipelines with Airflow or Prefect and expose APIs with Python/FastAPI.
• Deploy AI workloads on AWS Bedrock, SageMaker, Lambda, S3, Aurora/RDS, EC2.
• Work with Docker and Kubernetes to containerise workloads and ensure production readiness.
• Establish evaluation frameworks for LLM outputs and implement AI safety reviews.
• Collaborate with Data Engineers, Product Managers, and Tech Lead to translate business needs into AI solutions.
• Participate in Scrum ceremonies and promote best practices in AI engineering.

Conocimientos

LLM

RAG

multi-agent systems

Python

AWS

LangChain

LangGraph

Hugging Face

PyTorch

FastAPI

Docker

Kubernetes

Airflow

Prefect

Git

CI/CD

Terraform

Langfuse

Datadog LLM Observability

LangSmith

GitHub Actions

Graphs/Knowledge graphs

A2A/MCP protocols

Herramientas

LangGraph

LangChain

Hugging Face

PyTorch

Apache Airflow

Prefect

FastAPI

Docker

Kubernetes

Git

Terraform

SonarCloud

Snyk

GitHub Actions

Datadog LLM Observability

Langfuse

LangSmith

Descripción del empleo

We are looking for an AI Engineer to join a dynamic AI team within an international technology environment.

In this role, you will design, build, and deploy LLM-powered applications, RAG pipelines, multi-agent systems, and scalable AI solutions on AWS.

You will work at the intersection of software engineering, AI engineering, data pipelines, and cloud deployment, contributing to real use cases across the group. This role is highly hands-on and focused on building intelligent systems that move from experimentation to production.

You will work closely with Data Scientists, Platform and DevOps Engineers, Data Engineers, domain experts, Product Managers, and the Tech Lead to move AI solutions from experimentation into production.

If you enjoy working with LLMs, agents, RAG, vector databases, Python APIs, and AWS-native AI services, this could be a great fit.

What you’ll do

• Design, build, and maintain LLM-powered applications and multi-agent systems primarily using LangGraph and LangChain, alongside tools such as CrewAI or similar frameworks
• Develop and optimise RAG pipelines, including document ingestion, chunking strategies, embedding generation, retrieval logic, and vector search
• Implement and manage vector databases such as pgvector on Aurora, OpenSearch, Pinecone, or similar
• Build and maintain data and ETL pipelines using Apache Airflow, Prefect, or similar tools
• Develop backend services and APIs in Python / FastAPI to serve AI models, RAG systems, and agent workflows
• Deploy and manage AI workloads on AWS services such as Bedrock, SageMaker, Lambda, S3, Aurora/RDS, EC2
• Work with Docker and Kubernetes to containerise and orchestrate AI workloads
• Design and execute evaluation frameworks for LLM outputs, including automated testing, LLM-as-judge approaches, and human-in-the-loop review
• Work with LLM APIs and orchestration tools such as AWS Bedrock, OpenAI API, Anthropic API, or similar
• Apply prompt engineering and LLM evaluation methodologies, and assess when fine-tuning or other adaptation techniques are appropriate.
• Collaborate with domain experts, Data Engineers, Product Managers, and the Tech Lead to turn business requirements into AI solutions
• Participate in Scrum ceremonies and contribute to a collaborative Agile engineering culture
• Stay up to date with the rapidly evolving AI/ML ecosystem and proactively propose new tools, improvements, and approaches
• Support less experienced team members and share AI engineering best practices across the team.
• 3–5 years of experience in Software Engineering, with at least 1–2 years focused on AI / ML Engineering
• Strong proficiency in Python
• Experience with AI/ML and LLM frameworks such as LangChain, LangGraph, Hugging Face, PyTorch, or similar
• Hands-on experience building RAG systems, including embeddings, vector stores, semantic search, and hybrid search strategies
• Experience working with LLM APIs such as AWS Bedrock, OpenAI API, Anthropic API, or similar
• Solid understanding of prompt engineering, fine-tuning techniques, and LLM evaluation methodologies
• Hands-on experience with AWS services such as EC2, S3, Lambda, Aurora/RDS, Bedrock, SageMaker
• Experience with observability and evaluation platforms for LLMs such as Langfuse, Datadog LLM Observability, LangSmith
• Experience with Docker and Kubernetes
• Familiarity with data pipeline tools such as Apache Airflow, Prefect, or similar
• Experience developing backend services or APIs, ideally with FastAPI
• Proficiency with Git and software engineering best practices
• Experience working in a Scrum Agile environment
• Strong problem-solving, analytical thinking, communication, and teamwork skills
• Experience with multi-agent architectures and protocols such as A2A or MCP
• Familiarity with MLOps practices: model versioning, experiment tracking, MLflow, Weights & Biases, and CI/CD for ML
• Knowledge of graph databases or knowledge graphs for enhanced retrieval
• Experience with CI/CD pipelines using tools such as GitHub Actions
• Familiarity with Infrastructure as Code, especially Terraform
• Experience with code quality and security tools such as SonarCloud, Snyk
• Experience in aviation, travel, or large-scale digital environments
• Spanish language skills are a plus

Hybrid model - 2 days onsite per week

Why join this project?

• People first – diverse and inclusive culture in an international environment.
• Build production-ready LLM applications, RAG systems, and agentic AI solutions
• Work with cutting-edge AI technologies across the LLM, agents, vector search, and AWS ecosystem
• Contribute to scalable engineering practices around AI applications, data pipelines, evaluation, and deployment
• Gain hands-on exposure to AWS-native AI services such as Bedrock, SageMaker, Lambda, S3, and Aurora
• Be part of a fast-moving AI environment where experimentation, ownership, and impact are highly valued
• High team stability and collaborative culture.
• €1200 per year training budget and continuous learning opportunities.
• Private health insurance and benefits package.
• Flexible working hours and hybrid model.
• Wellhub: fitness, wellness, and mental health support.
• Football and paddle tennis teams sponsored by Capitole.
• Team buildings, global events, and strong tech communities.

Information Security Notice

The employee will have access to confidential information related to Capitole and the assigned project.

Compliance with internal security and information protection policies is mandatory.