Junior AI Engineer
Pubblicato il 15-10-2026 - Sdg Group in Milano
SDG Group is looking for Junior AI Engineers to join our AI & Data Science team.
From day one, you will be fully integrated into an international delivery unit, working with multicultural teams on high-impact projects across cutting-edge and highly competitive sectors such as Fashion & Luxury, Utilities, Banking, Insurance, and Healthcare.
This role is ideal for candidates eager to grow fast in a dynamic, international, and constantly evolving environment, with direct exposure to Generative AI and advanced analytics projects.
Your Challenges:
- Full-Lifecycle AI Development: Manage the entire journey of an AI solution, from initial problem-solving and data cleaning to building production-ready systems.
- Cutting-Edge Modeling: Implement a wide range of algorithms—from traditional statistical optimization and predictive modeling (supervised/unsupervised) to advanced Generative AI using LLMs and Transformers.
- Industrialization & AIOps: Transition models from local environments to the cloud using an engineering mindset.
You will ensure systems are scalable and reliable by applying AIOps methodologies.
- Cloud Mastery: Work across the "Big Three" (AWS, GCP, and Azure) and leverage specialized platforms like Databricks, SageMaker, and Vertex AI to handle large-scale data science needs.
- Strategic Insight & Communication: Translate complex datasets into actionable business insights. You will act as a bridge between technical data and client needs, presenting findings that solve real-world challenges.
- Collaborative Deployment: Work in synergy with Data Engineers to integrate models into live systems, ensuring they are not just "smart" but operational.
Requirements:
- Education: Master’s degree in a STEM field (Computer Science, Engineering, Mathematics, Statistics, Data Science, or related).
- Experience: Recent graduates are welcome, even with limited professional experience.
- Programming: Proficiency in R, Python and its ecosystem (NumPy, Pandas, scikit-learn).