An AI system to categorize semiconductor field failures
Overview
| Company Name / Department | NXP Semiconductors |
| Contact Person | Mohamed Moosa (based in Austin, Texas, USA) |
| Location | Eindhoven (anticipated location of the student) |
| Optional remote work | 3 days work from the office, 2 days possible from home in alignment with supervisor. |
| Travel expenses (own account or reimbursed by the company) |
In principle travel expenses are not applicable for students, as they have a student public transport (OV) card. If the student can proof that he/she has no student public transport card (DUO evidence is therefore needed) an ‘Altijd Vrij abonnement’ or RAV 100 incl. fiscal commuting allowance can be applicable for the student. More information about the RAV 100 The amount of the commuting allowance is € 0.19 per kilometer with a maximum of 35 kilometers for one-way commute. The distance is determined based upon of the number of kilometers between your home and work location, using the postcodes according to the ANWB route planner (shortest route). This travel allowance will be paid in accordance with applicable fiscal legislation. In addition to the above-mentioned commuting allowance, NXP offers the possibility of a net pay-out of the unused tax-exempt part of the commuting allowance (fiscal space travel allowance). This is settled with your gross salary. |
| Housing arranged by company |
No |
| Housing expenses (how much per month, own account or subsidized by the company) | If the distance between your residence and the residence where you will do your internship is more than 50 kilometers (from center to center), you can rent a room in Eindhoven. In this case you will receive a contribution in the rent of maximum € 400,- net a month. |
| Internship compensation | €600 per month |
| Study program | |
| Start date | September 1, 2026 |
Company Description
NXP Semiconductors N.V. is a Dutch semiconductor manufacturing and design company with headquarters in Eindhoven, Netherlands. It is the third largest European semiconductor company by market capitalization as of 2024. The company employs approximately 34,000 people in more than 30 countries. It designs and produces solutions in the Automotive, Industrial and Consumer markets that enable secure connected sensing, thinking and acting intelligently to improve people’s daily lives.
Project Description
With the rapid proliferation of semiconductor usage in the Internet of Things, field failures pose challenges to both semiconductor suppliers and users to rapidly determine the root cause of failures thus enabling fixes to enhance overall system availability and reliability. This project proposes the development of a machine learning and computer vision system that would be trained on an existing database of semiconductor field failures and their symptoms, that would subsequently be applied to newly observed field failures to infer their root cause of failure.
Goal of the Project
Combine natural language processing and image classification/segmentation using a subset of existing failure analyses reports to create an agent that can then be used to boot-strap a machine learning model for the much larger database of failures available within the company.
Deliverables
There are a few vectors that could be followed and the final set of deliverables will be agreed-upon in consultation with the student and mentor. In general, it’s expected that Dataiku (a python-based data science application) may be used for rapid prototyping of the NLP and CV capabilities.
Essential student knowledge
Machine Learning, Computer Vision (image classification and segmentation), neural networks, generative AI, natural language processing. A familiarity with CMOS circuits would be a plus.
More information: escf@tue.nl
