AI Strategy & Architecture for Alternance : Automated Test Program (TP) Generation F/M

Test Program (TP) development is a key activity within NXP Test Engineering, requiring high quality, consistency, and strict integration into established development workflows (code reviews, CI/CD, quality checks). Recent advances in Artificial Intelligence and Generative AI offer potential to automate and accelerate structured and repetitive parts of TP development. However, applying these technologies in an industrial context requires a clear strategy, well-defined architecture, feasibility assessment, and compliance with NXP processes and AI governance.

Apprentice Objective

The objective of this internship is to define an AI strategy and target architecture for automatic generation of Test Programs (TP) , with a focus on engineering feasibility rather than full implementation .

The apprentice will:

• Propose a modular architecture for AI-assisted TP generation
• Assess feasibility and risks for each architectural block
• Identify a realistic and compliant minimum viable scope
• Optionally develop a proof of concept for one selected block to validate the approach

A key design constraint is usability:

The proposed solution should aim for "no setup dependencies" for end users (easy to deploy, reproducible, minimal environment configuration).

Scope of Work

The internship activities will include:

• State-of-the-art review of AI and Generative AI approaches applicable to code and test generation
• Definition of a target architecture , including:

• functional blocks and responsibilities
• data flow and interfaces
• integration into existing TP development workflows

• Feasibility assessment per block , covering:

• required inputs and dependencies
• technical risks and limitations
• validation and quality considerations
• integration and governance constraints

• Definition of recommendations and next steps
• Optional implementation of one architectural block or a limited end-to-end slice as a proof of concept

Expected Deliverables

• State-of-the-art summary (focused on applicability to TP generation)
• Architecture proposal (block diagram and description)
• Feasibility matrix per block (Go / Conditional / No-Go assessment)
• Optional proof of concept for one selected block
• Final recommendations and roadmap proposal

• Master's or Engineering student in Computer Science, Software Engineering, AI, or related field
• Strong software engineering fundamentals (Python and/or Java)
• Interest in AI, automation, and software architecture
• Structured mindset and ability to document and communicate technical findings

Nice to have:

• Knowledge of CI/CD, code quality, or test automation concepts
• Familiarity with Generative AI / LLM principles