Keywords: Machine Learning, Datasets, Anomaly detection, Production system.
At Motorola Solutions we are the global leader in innovative mission-critical communication solutions. Keeping leadership requires continuously adoption of new technologies to leverage the endless possibilities. We build cloud services that are used by police officers worldwide and our DevOps have to ensure our cloud services are operational 24/7, 365 days a year.
The first step towards making a system build out of many computer programs self-healing is to automate analysis of the ever-growing amount of logs and metrics produced by modern software.
What we provide for your research:
- A dataset with logs and metrics from a production system
- Documented customer issues from the past that can be correlated against the logs and metrics (e.g. caused by end-user loaded wrong certificate, system-admin applied wrong configuration, dependency broke down)
- A team of experts (who also developed the application) that helped root-cause and resolve these issues, and that can help you to correlated issues and logs/metrics
Example of thesis questions:
- Which filtering techniques give the best results with training a machine learning model to detect anomalies in a telecommunication system?
- Which anomalies can an expert system detect in a telecommunication system, and which anomalies cannot be detected by an expert system?
- Which input gives the best results for anomaly detection with machine learning; logs, metrics or traces?
The ideal candidate
You are a master student in computer science or similar. You have theoretical or practical experience in deploying Machine Learning models. Experience in DevOps methodologies like Docker and Kubernetes. Understand REST APIs and RabbitMQ message broker. You are ambitious with a good eye for details and keen on making this project a success. We expect you to be ambitious and have a strong interest in software development. You are able to work independently and focused and you don’t let challenges get in the way of reaching your goals. You also have good communication skills in Danish or English.
If you are interested in the project, apply here or send an e-mail email@example.com.