Security Research Intern

Your tasks

We are looking for highly motivated students to join our research team in developing solutions that apply machine learning/deep learning algorithms and formal methods to security and privacy problems.

Example applications include, but are not limited to, utilizing AI and formal methods to analyze software vulnerabilities, modeling of secure systems, automation of security engineering processes, security of generative models, and fine-tuning models for specific security use cases.

This internship role will focus on the following responsibilities:

• Design and development of novel methods/systems, implementation, and evaluation using benchmarks and Bosch data from various security application domains.

• Technical discussions with the research team towards creation of new ideas and improve upon the existing solutions.

• Clean, efficient, and well-documented implementation of new methods.

Contributions to Bosch IP portfolio.

Your profile

Required:

• Pursuing PhD degree in Computer Science / Engineering, Electrical Engineering, or Mathematics.

• Demonstrable experience with applying LLMs and ML/DL techniques to solve real-world problems.

• Proficient in programming (e.g., Python, C, C++, Java) and use of AI frameworks and libraries (e.g., PyTorch, scikit-learn).

• Experience with LLMs, model fine tuning, and alignment.

• Excellent communication skills.

Desired:

• Experience with system security, software vulnerabilities, static code analysis methods, formal methods (e.g., SMT solvers).

• Established publication record.

• Good coding practices, ability to write efficient and readable/maintainable code.

• Good writing and presentation skills.

Contact & additional information

Equal Opportunity Employer, including disability / veterans.  

Bosch adheres to Federal, State, and Local laws regarding drug-testing. Employment is contingent upon the successful completion of a drug screen and background check. Candidates who have been offered the position must pass both screenings before their start date.