Brembo Hackathon

At the heart of the hackathon lies a vast and diverse dataset provided by Brembo. This dataset encompasses a wealth of information on various car models, their braking systems, performance metrics, and real-world driving scenarios. Armed with this invaluable data, participants will be tasked with building innovative machine learning models capable of predicting and optimizing brake designs, or, solving warranty database issues to further improve Brembo’s already in-place model.

Participants will choose between one of two hackathon challenges

which are defined below. Each challenge will have it’s own unique data set, sample modeling, and will have a workshop/webinar to be held

1-2 weeks before the hackathon to explain how the data is structured. Register today and don’t miss out on this exciting opportunity to work with automotive experts and transform the next generation of Machine Learning and AI with Brembo!

Using friction test data provided by Brembo, use Generative AI to create new compounds, forecast testing results and create the framework for predicting the effectiveness and characteristics of a new Brembo brake product. The data provided will include a list of compounds previously used and tested by Brembo, as well as their outcomes. Solutions must be based on Generative AI, applied to provide a model able to propose new recipes that increase the number of candidate compounds, ensuring feasibility and good performances.

Submission Requirements:

Dataset of proposed compounds in the std. format provided by Brembo

Presentation Deck

Reproducible source code and related documentation (this will not be required but strongly recommended so the judges can properly score solutions)

Brembo will then evaluate the proposed compounds and give feedback to the teams in terms of:

Technical Feasibility: Validation of proposed compounds against production constraints.

Technical Relevance: Comparison of test conditions (specific columns) against reality.

Target Performance: Comparison of the proposed compound’s performance against the target.

Compounds Variability: Computation of a variability index among all the proposed compounds.

Submission Requirements:

Synthetic dataset in the std. format provided by Brembo

Presentation Deck

Reproducible source code and related documentation (this will not be required but strongly recommended so the judges can properly score solutions)

Brembo will then evaluate each team's submission and:

Train the target model on the augmented dataset, obtaining a newly trained model.

Compare performances of the newly-trained model with the target model on the test set.

Provide feedback to the teams in terms of mean per-class error.

Registration Opens!

Webinar 1: Join us for the first webinar w/ Brembo explaining Challenge 1.

Webinar 2: Join us for the second webinar w/ Brembo explaining Challenge 2.

Hackathon Pre-Kick-off: Join us for the pre-kick-off webinar to learn more about the event!

Hackathon Start Date: The hackathon officially starts October 13th at 6:00 pm!

Hackathon Day 2: Participants continue working on-site for day 2.

Hackathon Day 3: Projects will be due, judges will review the projects and winners will be announced!

Brembo leads the world in the design and production of high-performance braking systems and components for top-flight manufacturers of cars, motorbikes and commercial vehicles. Founded in 1961 in Italy, Brembo has a long-standing reputation for providing innovative solutions for OEMs and aftermarket. Brembo also competes in the most challenging motorsport championships in the world and has won over 600 titles. Guided by its strategic vision – “Turning Energy into Inspiration” – Brembo’s ambition is to help shape the future of mobility through cutting-edge, digital and sustainable solutions. With about 15,000 people across 15 countries, 30 production and business sites and 8 R&D centers, Brembo is the trusted solution provider for everyone who demands the best driving experience.

Hackathon powered by

Check out these helpful resources