Internship in agentic AI for automated code generation in safety-critical embedded software

Internship at Bosch Rexroth in Schwieberdingen on agentic AI for automated code generation in safety-critical embedded software. Enrollment at a university is required. Duration: 6 months, start by agreement.

• During your internship you analyze our current development process (requirements, coding, unit tests, reviews and integration) and identify the steps with the highest automation potential.
• You evaluate state-of-the-art agentic coding tools and frameworks (e.g. Claude Code, GitHub Copilot Agent Mode, MCP-based toolchains, multi-agent orchestration) for use with C/C++ on QNX and embedded Linux.
• You also develop a process for building an agent pipeline that takes a requirement or ticket and generates a pull request. This includes code, unit tests, static analysis (MISRA C/C++, clang-tidy), and builds and tests in CI with self-correction loops.
• You also familiarize the agents with our context using a structured knowledge base from the repository, architecture documentation and coding guidelines, e.g. via RAG, project rules and context engineering.
• You work on addressing safety and security requirements, e.g. by ensuring traceability to requirements, performing human-in-the-loop reviews and ensuring compatibility with ISO 13849/IEC 61508 and CRA-driven processes.
• From day one you measure the impact using clear KPIs, e.g. lead time per ticket, test coverage, error rate, review findings and cost per feature. These are compared with a manual baseline.
• Last but not least, you deliver a working prototype, guidelines for the teams in Germany, Poland and India, and a recommendation for the rollout.

• Education: Master's studies in computer science, software engineering, electrical engineering, information technology, mechatronics or similar
• Experience and knowledge: hands-on work with LLMs and AI coding assistants (e.g. Claude Code, GitHub Copilot)
• Experience and knowledge: experience with Git, CI/CD (e.g. GitHub Actions, Jenkins) and Linux
• Experience and knowledge: very good Python skills, good working knowledge of C or C++
• Experience and knowledge: first practical experience in software development (e.g. through an internship, a working student job, university projects or personal projects – GitHub link welcome)
• Experience and knowledge: ideally first projects with LLMs beyond plain chat use
• Personality and working style: you work independently, like to experiment and present results clearly and critically
• Work routine: your presence on site is required
• Languages: fluent English; good German skills are an advantage