Data Scientist level I - internship

About us

Presto Engineering is the recognized expert in ASIC design and integrated semiconductor services with over 35 years of experience in the industry. We help innovative companies imagine and implement microelectronic solutions for tomorrow, in high-value markets such as Medical, Automotive, Industrial and Communication, while reducing risk, minimizing fixed cost structure, and accelerating time-to-market.

About The Job

This is a 6-month internship.

Reporting directly to the local Product Engineer, the Data Scientist Level I position is responsible for developing and improving a classifier/model dedicated to detecting different abnormal behavior in the production based on the main logs and wafer map images. The intern may involve different tools to evaluate, visualize and track the improvement of the model.

Your Responsibilities

Main responsibilities

• Analysis of the existing model: Understand the current AI architecture, its training process, and its performance on silicon production images.
• Dataset analysis: Explore the available image dataset, assess image quality, identify class imbalances, and evaluate the quality of annotations.
• Performance evaluation: Measure the model's performance using relevant metrics, including accuracy, precision, recall, F1-score, and confusion matrices.
• Error analysis: Identify recurring misclassifications and investigate their potential causes, such as visual similarities between classes, image noise, or variations in acquisition conditions.
• Model improvement: Experiment with different approaches, including data augmentation, hyperparameter optimization, transfer learning, and alternative model architectures.
• Robustness assessment: Evaluate the model's ability to generalize to new images and change production conditions.
• Performance optimization: Investigate opportunities to reduce inference time and computational requirements while maintaining classification quality.
• Documentation: Summarize the experiments, compare the results, and formulate recommendations for further model development and potential industrial deployment.

Secondary responsibilities

• Quality: Respect quality process and documentation in test company policy
• Safety & Environment: Respect safety and environment company policy
• Security: Respect the Security company policy
• IT & Technologies: Respect IT & Technologies policy

Requirements

• Currently pursuing a Master's degree or an engineering degree in Artificial Intelligence, Computer Science, Data Science, Computer Vision, or a related field.
• Solid programming skills in Python.
• Basic understanding of machine learning and deep learning concepts.
• Knowledge of image classification and computer vision techniques.
• Ability to analyze experimental results, identify model limitations, and propose improvements.
• Strong analytical and problem-solving skills.
• Ability to work independently and collaboratively within a technical team.
• Good communication skills and the ability to document technical work clearly

Technical skills

• Programming: Python
• Deep learning frameworks: PyTorch or TensorFlow/Keras
• Computer vision: OpenCV, image preprocessing, and image augmentation
• Data analysis: NumPy, Pandas, Matplotlib, and Seaborn
• Model evaluation: Scikit-learn and relevant classification metrics
• AI methods: Convolutional Neural Networks (CNNs), transfer learning, and potentially Vision Transformers (ViTs)
• Development tools: Jupyter Notebook, Git, and standard machine learning workflows

Location & Travel

The role will be based at our office in Meyreuil, France.

How to apply

Please send us your application and CV in English only by selecting the 'Apply Now'.

Want to Know More?

To know more about us, please visit our website to find out what it's like to work at Presto Engineering, and follow us on LinkedIn for updates.