Internship Quantization Of State-Of-The-Art Nn Models Performing Semantic Segmentation Torino

Internship Quantization Of State-Of-The-Art Nn Models Performing Semantic Segmentation
Torino
Piemonte, Torino

OUR STORY

At ST, we believe in the power of technology to drive innovation and make a positive impact on people, business, and society. We are a global semiconductor company, and our advanced technology & chips forms the hidden part of the world we live in today.  

When you join ST, you will be part of a global business of more than 115+ nationalities and present in 40 countries, 50, 000+, diverse and dedicated creators & makers of technology around the world!

Developing technologies takes more than talent: it takes amazing people who understands collaboration and respect. People with passion and desire to disrupt the status quo, push boundaries and drive innovation – whilst unlocking your own potential.

Working at ST means innovating for a future that we want to make smarter, greener, in a responsible and sustainable way. Our technology starts with you. Join us and start the future!

YOUR ROLE 

The objective of this activity will be the quantization of state-of-the-art NN models performing Semantic Segmentation.

This activity is in the frame of the IMC Corporate Project and will be a continuation of the work carried out by other interns, for deploying NN models to ST platforms featuring NN HW accelerators.

There are already some NN candidates performing Semantic Segmentation to be mapped to the STM32N6 and possibly to platforms containing IMC blocks.

The activity is divided in five tasks:

Complete port to STM32N6 of the NN models that have been identified and modeled in QKeras

In case of performance degradation, improve the models that have been identified / identify other models

Complete this port to implement a real live demo with camera and display

(Optional) Port the same NN / a similar NN to a platform featuring IMC.

(Optional) Improve the already identified methodologies and techniques for optimizing the search of the optimal quantization parameters

YOUR SKILLS & EXPERIENCES

Minimum requirements

Master degree in Computer Science or Data Science or Statistics (or equivalent experience)

Knowledge of programming

Proven experience with ML, Deep Learning, Tensorflow, Python, NLP

Data science and statistics background

Knowledge of basic algorithms, object-oriented and functional design principles, and best practice patterns

Proficient in using Python and one or more major AI platforms

Good understanding of machine learning/AI Algorithms modeling and evaluation

Experience with STM32 platforms

Excellent mathematical skills

Fluency in English for effective communication

Preferred skills

Good organizational and communication skills

Experience with ST Cube framework

The Employment search is addressed to candidates of both genders, under Law 10. 04. 1991 n. 125, as amended by Legislative Decree n. 198/2006 which guarantees gender equality at work

We encourage candidates who may not meet every single requirement to apply, as we appreciate diverse perspectives and provide opportunities for growth and learning. Diversity, Equity and Inclusion (DEI) is part of our company culture. Our DEI vision is, "At ST, you can be the true version of yourself", we value all employee contributions and have zero tolerance for any kind of discrimination.

Joining us is also about a greater work-life balance and workplace with equal opportunities. Dedicated Employee Resource Groups for women and LGBTQIA+, hybrid work arrangements are amongst the many DEI & Sustainability initiatives that make us a great place to evolve your career.

To discover more, visit st.com/careers