
AI Engineer - Internship (Open also to Protected Categories, Law 68/99)
Job Description
Job title: AI Intern
As part of our digital transformation, we are launching an innovative Proof of Concept to develop an AI-assisted tool capable of automating the initial drafting of an FMEA. The AI should analyze unstructured and structured exports (e.g. BOMs) and cross-reference them with historical reliability files to predict failure modes and their local effects.
Key Responsibilities
In the Operations team, you will:
- Analyze electronic data: understand and parse unstructured (pdf drawings) and structured outputs from standard EDA software (e.g., Mentor Graphics, Cadence).
- Develop an AI/Data pipeline: design and implement a Python-based AI strategy, combining NLP and Computer Vision techniques (LLMs and LVMs) with graph-based approaches to map circuit topologies and understand component relationships.
- Integrate historical data: connect your algorithm to historical failures files to accurately predict how specific components (e.g., resistors, capacitors, ICs) fail within their specific circuit context.
- Generate FMEA reports: structure the AI’s output into a standardized FMEA format that human engineers can review and validate.
- Test and validate: Work closely with hardware engineers to validate the AI's logic on a simple, baseline electronic board to prove the concept's viability.
Requirements:
- Currently pursuing a Master’s degree or in the final year of an engineering School. Preferred majors: Electrical Engineering or Computer Science /AI.
- Strong programming skills in Python.
- Experience with AI/Machine Learning concepts (NLP, LLM prompting/fine-tuning, or data structuring).
- Fundamental understanding of electronic circuits, components, and schematics.
- Familiarity with EDA tools and file formats (Netlists, BOMs) is a plus.
- Basic knowledge of reliability engineering concepts or FMEA is advantageous.
Soft Skills
- Hybrid Thinker: Ability to bridge the gap between hardware engineering and software/AI development.
- Problem Solver: A pragmatic approach to scoping AI projects (starting simple and scaling up).
- Autonomous & Curious: Eager to dive into historical technical data and experiment with new algorithmic approaches.
- Good communication skills to present your findings to both software and hardware teams.