Altia is a leading Spanish digital transformation and IT consulting firm that provides end-to-end technology services, including software development, cloud computing, and cybersecurity. Founded in 1994 and listed on the BME Growth market, the company has grown into a major player in the Iberian technology sector with a workforce of over 3,200 professionals.

Headquartered in A Coruña, Altia maintains a strong nationwide presence with offices in Madrid, Barcelona, Bilbao, and several other Spanish cities. The company is particularly relevant for international professionals due to its status as a major contractor for European Union agencies and its diverse range of technical roles spanning Artificial Intelligence, Data Science, and enterprise software development.

Founded in 1994 in A Coruña, Altia began as a local IT services provider before expanding nationally across Spain. In 2010, it became a public company listed on the BME Growth market, and in 2019, it significantly expanded its international footprint through the acquisition of the Portuguese firm Noesis.

AI Engineer

Altia

This job was originally posted in Spanish and automatically translated to English. You'll most likely need Spanish to apply.

Job Description

We are a team of around 4,000 professionals dedicated to creating future-ready digital solutions that generate real value and drive significant change.

With a presence in more than 25 locations across 9 countries, we bring our innovative vision to projects for more than 1,000 organizations, helping them transform and grow sustainably.

For 30 years, our purpose has been to grow by helping others grow, evolving alongside our clients to fully seize opportunities and build a better world together.

📌 Project Summary:

We are looking for an AI Engineer for a strategic project within a public body: the analysis, design, and implementation of an on-premise data center (CPD) for the development of AI solutions without dependencies on public clouds and under a Green AI approach.

You will join a benchmark technical team, working with open source and open weights models in 100% on-premise scenarios. You will have a direct impact on architecture decisions, evaluation and fine-tuning of models (generative and non-generative based on transformers), as well as the creation of SLMs from LLMs.

🔑 Key Responsibilities:

* On-prem AI Solution Design: Define and implement end-to-end pipelines for training, evaluation, inference, and operation of AI solutions in an internal data center, maximizing energy efficiency and sustainability.
* Model Evaluation and Selection: Identify, compare, and evaluate open weights models (generative and encoder) using objective metrics and reproducible benchmarks.
* Fine-tuning and Optimization: Adjust transformer models (generative and non-generative) and apply light adaptation techniques (e.g., LoRA/Adapters) and optimization (distillation, quantization) where appropriate.
* LLMs and RAG: Design and implement RAG architectures, prompt and context engineering, embeddings, and semantic retrieval with vector databases.
* Integration and APIs: Expose capabilities via REST APIs (FastAPI or similar), ensuring quality, observability, and security in on-premise and containerized environments.
* Engineering and Quality: Apply best practices for development, testing, maintainability, traceability, and technical documentation; contribute to internal guides and peer reviews.
* Collaboration and Knowledge: Work within Scrum frameworks, share knowledge with the team, and contribute to the technical evolution of the data center and Green AI practices.

🎓 Requirements:

* Experience: 2+ years executing AI/ML projects with similar technologies in production environments.
* Languages and Development: Proficiency in Python applied to AI and data solutions; use of Pandas.
* Generative AI and NLP: Experience with LLMs (encoder and generative), prompt and context engineering, and RAG architectures.
* Frameworks and Ecosystem: PyTorch and Hugging Face.
* APIs and Integration: Experience building and integrating REST APIs with FastAPI (or equivalent open source frameworks).
* Open Weights Models: Experience in the selection and evaluation of open source models.

🛠️ Valued Technical Competencies:

* Model Adaptation: LoRA, Adapters, distillation, and quantization.
* Platform and Data: Kafka; MinIO (or similar) for storage; Kubernetes for orchestration.
* Use Cases: Computer Vision, time series, and classical ML.
* Cloud: Knowledge of AWS, Azure, GCP, or OCI (for hybrid scenarios or portability).
* Engineering Practices: MLOps, monitoring, CI/CD, observability, and security in on-premise environments.
* Certifications: Scrum, public clouds, or other related certifications.

This job offer has been written taking into account impartiality and non-discrimination based on gender, race, ideology, or any other reason. Specifically, respect for current regulations regarding equality between women and men (Royal Decree-law 6/2019) has been considered.

View original advert (Spanish)

Somos un equipo de alrededor de 4.000 profesionales dedicados a crear soluciones digitales preparadas para el futuro que generan valor real e impulsan un cambio significativo.

Con presencia en más de 25 ubicaciones en 9 países, llevamos nuestra visión innovadora a proyectos para más de 1.000 organizaciones, ayudándolas a transformarse y crecer de manera sostenible.

Durante 30 años, nuestro propósito ha sido crecer ayudando a otros a crecer, evolucionando junto a nuestros clientes para aprovechar plenamente las oportunidades y construir juntos un mundo mejor.

📌 Resumen del Proyecto:

Buscamos un/a Ingeniero/a IA para un proyecto estratégico en un organismo público: análisis, diseño y puesta en marcha de un CPD on‑premise para el desarrollo de soluciones de IA sin dependencias de nubes públicas y bajo un enfoque de IA verde.

Te incorporarás a un equipo técnico de referencia, trabajando con modelos open source y open weights en escenarios 100% on‑premise. Tendrás impacto directo en decisiones de arquitectura, evaluación y ajuste fino de modelos (generativos y no generativos basados en transformer), así como en la creación de SMLs a partir de LLMs.

🔑 Responsabilidades Clave:

* Diseño de Soluciones IA On‑prem: Definir e implementar pipelines end‑to‑end para entrenamiento, evaluación, inferencia y operación de soluciones IA en CPD propio, maximizando eficiencia energética y sostenibilidad.
* Evaluación y Selección de Modelos: Identificar, comparar y evaluar modelos open weights (generativos y encoder) con métricas objetivas y benchmarks reproducibles.
* Fine‑tuning y Optimización: Ajustar modelos transformer (generativos y no generativos) y aplicar técnicas de adaptación ligera (p. ej., LoRA/Adapters) y optimización (destilación, cuantización) cuando proceda.
* LLMs y RAG: Diseñar e implementar arquitecturas RAG, ingeniería de prompts y de contexto, embeddings y recuperación semántica con BBDD vectoriales.
* Integración y APIs: Exponer capacidades mediante APIs REST (FastAPI o similares), asegurando calidad, observabilidad y seguridad en entornos on‑premise y containerizados.
* Ingeniería y Calidad: Aplicar buenas prácticas de desarrollo, testing, mantenibilidad, trazabilidad y documentación técnica; contribuir a guías internas y revisión por pares.
* Colaboración y Conocimiento: Trabajar en marcos Scrum, compartir conocimiento con el equipo y contribuir a la evolución técnica del CPD y la práctica de IA verde.

🎓 Requisitos:

* Experiencia: 2+ años en ejecución de proyectos de IA/ML con tecnologías similares en entornos productivos.
* Lenguajes y Desarrollo: Dominio de Python aplicado a soluciones de IA y datos; uso de Pandas.
* IA Generativa y NLP: Experiencia con LLMs (encoder y generativo), prompt y context engineering, y arquitecturas RAG.
* Frameworks y Ecosistema: PyTorch y Hugging Face.
* APIs e Integración: Experiencia construyendo e integrando APIs REST con FastAPI (o frameworks open source equivalentes).
* Modelos Open Weights: Experiencia en selección y evaluación de modelos open source.

🛠️ Competencias Técnicas Valoradas:

* Adaptación de modelos: LoRA, Adapters, destilación y cuantización.
* Plataforma y datos: Kafka; MinIO (o similar) para almacenamiento; Kubernetes para orquestación.
* Casos de uso: Computer Vision, series temporales y ML clásica.
* Cloud: Conocimientos de AWS, Azure, GCP u OCI (para escenarios híbridos o portabilidad).
* Prácticas de Ingeniería: MLOps, monitorización, CI/CD, observabilidad y seguridad en entornos on‑premise.
* Certificaciones: Scrum, nubes públicas u otras relacionadas.

Esta oferta de trabajo ha sido redactada teniendo en cuenta la imparcialidad y la no discriminación por razón de género, raza, ideología o cualquier otra razón. Específicamente, se ha considerado el respeto a la normativa vigente en materia de igualdad entre mujeres y hombres (Real Decreto-ley 6/2019).