Enthusiastic and motivated mathematicians with fundamental programming skills are ready for good standards and high quality

06/12 – now BMW AG, Munich
• Deep Learning: Pattern recognition by CNN in multidimensional trajectories
o Autoencoder: Bird Eye View
o Hierarchical clustering on n-dimensional vector space
• Supervised Learning: Learning of machine algorithms for
o Detection of override maneuvers (SVM, KNN, decision tree)
o Prediction of the code generation time of Simulink models (linear regression)
• Reinforcement Learning: First steps in
o Walking roboter (model based example), Simulation of reinforcement learning agent
• AWS: EC2, Lambda functions, S3, SageMaker, SQS, SNS, IAM, Cloudformation
• Tool development for automatic C code generation from simulation models
• Optimization of C code generation from Simulink models
• Unit testing in Matlab
• Analysis of driving maneuver measurements
• Set up/manage Jenkins jobs
• Implementation of functions for automatic model generation
• Create Docker Container for model-driven software development
• Supervision of Master and Bachelor theses, as well as interns

07/10 – 05/12 Bertrandt AG, Munich
• Model-based software development of production software in the field of vehicle dynamics
• Commissioning of software on the vehicle
• Debugging and bugfixing of software modules
• Versioning of software artifacts in CM-Synergy and Subversion
• Flashing of control units and integration of software versions (commissioning)

10/03 – 03/10 Graduate mathematician, university

Work Terms

- Average working hours per week are between 10 and 20 hours
- Current situation only allows remote activity
- Payment should be made at the end of the project. For projects longer than two months individual payments should be agreed upon.
- Contact is possible: by phone, Skype and e-mail