STAGE - Qualité de conception GEN7 (F/H)

Dassault Systèmes is a major player in industrial software. Its 3DEXPERIENCE platform allows its customers to design, simulate, and produce real-world objects in a virtual universe, from aeronautics to healthcare.

Within CATIA R&D, the Sustainability team designs and develops the platform's design and engineering indicators. You will be integrated into this organization, within the Engineering Quality department.

This department develops applications that allow manufacturers to formalize their business rules - checks, profiles, design rules - and then automatically verify the compliance of their digital models. These applications produce rich quality reports: overall score, severity and status per rule, list of defaulting elements, annotations, screenshots, associated correction functions.

However, their interpretation remains reserved for experts. An engineer who opens a report must still understand alone why a rule has failed, and decide what to do next. This missing link is precisely what this internship addresses.

As part of Gen7 and the AI assistant embedded in the platform, the team is building a new generation of conversational tools. Your work will be an integral part of this.

Your missions

Your subject: design and develop a tool capable of analyzing a quality evaluation result, explaining its defects in natural language, and deducing appropriate corrective actions - up to automatic correction (healing).

Specifically, you will be required to:

• Study the existing result model - score, weight, status per rule, defaulting elements, correction functions - as well as the documentation for some 500 rules delivered as standard;

• Design the analysis logic: group and prioritize defects by severity and impact, identify recurring causes, and distinguish what can be automatically corrected from what requires human arbitration;

• Generate, using a language model, a clear explanation of each defect - the why and how - anchored in the product documentation to avoid any fabricated response;

• Propose and then chain corrective actions: execution of a healing function, documented justification, or manual review - with explicit user confirmation before any modification of the model;

• Integrate everything as a tool declared in the platform's AI infrastructure: JSON declaration, C++ implementation, result contract;

• Set up a method for evaluating the quality of the responses produced: sets of reference conversations, measurement of regressions, analysis of failure cases.

You will work in direct conjunction with the domain architects and the AI infrastructure team. The subject is open: your design choices will be discussed, and what you produce is intended to be incorporated into the product.

Your qualifications

Education

Student in the final year of an engineering school or a Master 2 in computer science, specializing in software engineering with a focus on artificial intelligence.

Technical skills

• Solid foundation in software development; C++ practice is a real asset, Python and JavaScript are appreciated;

• Strong interest and ideally first experience with language models: prompt design, tool calling, research-augmented generation (RAG);

• Taste for data modeling and API design;

• Fluent technical English, both written and oral - documentation and daily exchanges with international teams.

Personal skills

• Analytical mind and ability to formalize business reasoning into exploitable rules;

• Autonomy and curiosity: the subject is open and you will be proactive;

• Rigor - a recommendation produced must be accurate and verifiable: an erroneous advice on a model has a real cost for the client;

• Ease in communicating and discussing your design choices with the team.

Joining us also means

Joining a scientific company at the heart of technological innovation, driven by strong growth for over 40 years.

Main advantages and benefits:

• Collaborative and innovative environment

• International collaboration

• Diversity of technologies, products and solutions

• Commitment to diversity and inclusion