Job opening

Deep Learning Engineer Dynamic Neural Networks (m/f/d) company follow us . company follow us .

Axelera AI /  Eindhoven (NL)

Apply on site

Job description

Do you get excited by working on cutting-edge Computer Vision models to bring Artificial Intelligence applications to the next level? We are expanding our team with a Deep Learning Engineer Dynamic Neural Networks to optimize the latest deep neural networks for our state-of-the-art neural-network accelerator chip. Your work will bridge the gap between various Machine Learning applications (security camera, drone, retail, robotics, etc.) and deep learning acceleration.

Dynamic Neural Networks aim at making the common (easy input) case fast (efficient), by trying to only enable parts of the neural network that are relevant given the current input data, leaving the rest dormant. This is realized by building decision blocks into the network, which decide to execute or skip certain subsequent layers. The decision blocks are trained together with the network weights. This has the advantage that at run-time the activity (and hence power consumption) is scaled with the complexity of the incoming data, enabling some form of event-driven operation. The result is a faster and more efficient inference of state-of-the-art neural networks

This position will start in October or November 2022.

Your role

Your role will include the following:
* Thoroughly studying various Dynamic Neural Network techniques
* Implementing and training existing Dynamic Neural Network algorithms
* Developing new Dynamic Neural Network algorithms to further accelerate Deep Learning algorithms on our accelerator
* Integrating Dynamic Neural Networks into the compiler flow
* Keeping track of the literature on Dynamic Neural Network
* Collaborating with software developers
* Communicating your results to team members

Employer

Axelera AI is a truly European deep-tech Start Up company which is developing a game-changing hardware and software platform for AI at the edge that will make the industry more integrated, efficient and accessible. Our mission is to spread artificial intelligence for a green, fair, trusted and safe world enabling new application of AI in diverse sectors like smart cities, retail and other markets. Our company is a spin-off from a multinational deep-tech group and is backed by a strong syndicate of institutional investors. We have an extraordinary and international team of top talented researchers and developers working in the headquarter in Eindhoven (NL) and in the branch offices in Leuven (BE), Zurich (CH) and Pisa (IT).

Candidate/profile

* At least 4 years of experience with Python (Numpy, Scipy)
* At least 3 years of experience with a deep learning framework like Pytorchor Tensorflow
* Strong understanding of the inner-workings and applications of recent CNNs, MLPs, RNNs and transformers
* Experience with one or more Dynamic Neural Network techniques
* Preferably experience with Torch FX
* Preferably experience with Neural Network compilers
* Problem-solving mindset, capable of debugging and patching software issues
* Good oral and written communication skills
* Fluent in English, both in speaking and writing
* You are a team player and you are also able to autonomously plan and perform research tasks
* You have a strong sense of responsibility and want to realize high ambitions.

Conditions and benefits

Take the chance to become part of a dynamic, fast-growing, international organization. We offer an attractive compensation package, including a pension plan, extensive employee insurances and the option to get company shares. An open culture that not only supports creativity and continual innovation is awaiting you. Collaborative ownership and freedom with responsibility is characteristic for the way we act and work as a team.

Application procedure

Great! We are looking forward to receiving your application! Feel free to contact us whenever you have any questions or for further information @Heike Wilfling, HR Business Partner +31621929159. For further information on Axelera AI please also have a look on our website: www.axelera.ai, * Email: info@axelera.ai, * Email: info@axelera.com

Full description & Apply on site