Senior Agentic Software Engineer

Bosch Group•Northville Charter Township, MI

About The Position

ETAS’s Vehicle Information Solutions (VIS) organization is seeking an experienced and innovative Senior Agentic Software Engineer to develop and deploy production-grade AI systems. The role involves orchestrating Python-based back-end architecture, integrating large language models, optimizing vector databases, and deploying scalable, secure APIs on Microsoft Azure. The engineer will work closely with automotive OEM customers to agree on requirements, plan solution delivery using Agile processes, and uphold software quality standards throughout the development lifecycle. The position owns problems end to end, from understanding user needs through architecture, implementation, verification, and operation. Implementation is delivered primarily through AI coding agents: the engineer translates intent into precise specifications, constraints, and acceptance criteria, supervises multiple agent work streams, and applies human judgment to architecture, trade-offs, edge cases, and review. The role is accountable for the correctness, security, and maintainability of merged work and emphasizes tests, typed interfaces, CI gates, and evaluations to verify agent-produced code. Work may span frontend, backend, infrastructure, DevOps, and data, while development workflows evolve alongside advances in models and agent capabilities. VIS supports automotive OEMs and dealer networks with service information, wiring diagrams, technical knowledge, and digital aftersales solutions. It is transforming into a Vehicle Intelligence partner, applying AI, automation, and automotive expertise to support software-defined vehicles and future workshop experiences.

Requirements

Thorough understanding of software engineering fundamentals, including architecture, system design, testing strategy, and debugging.
Infrastructure and DevOps competence, including cloud platforms, CI/CD pipelines, containerization, infrastructure as code, and operating built systems.
Excellent communication skills, with the ability to express intent, requirements, and constraints clearly in writing and discussion.
Proven ability to build relationships with customers.
Strong abstraction skills, including decomposing complex problems, selecting appropriate levels of abstraction, and recognizing patterns that generalize across domains.
Hands-on production experience with agentic coding tools such as Claude Code or GitHub Copilot agents, including running multiple agents in parallel and unattended agents in automated pipelines.
Product thinking and ability to translate user and business intent into specifications precise enough for autonomous execution.
Prompt and context engineering skills, including writing instructions, specifications, and repository conventions that agents reliably follow.
A strong verification mindset: treating generated code as untrusted until independently checked.
Adaptability to unfamiliar languages, frameworks, and domains.
Security awareness for AI-assisted development, including secret handling, prompt injection, and supply-chain risks.
Bachelor’s degree in Software Engineering, Computer Science, or a similar field of study.
Indefinite U.S. work authorization; future sponsorship for work authorization is unavailable.
Successful completion of a drug screen and background check before the start date.

Nice To Haves

Customer relationship-building experience in the automotive industry.

Responsibilities

Own problems end to end, from understanding user needs through architecture, implementation, verification, and operation.
Deliver implementation primarily through AI coding agents by translating intent into precise specifications, constraints, and acceptance criteria, and supervising multiple agent work streams in parallel.
Make architecture decisions, assess trade-offs and edge cases, and critically review agent output; remain accountable for the correctness, security, and maintainability of merged work.
Design for verifiability using tests, typed interfaces, CI gates, and evaluations that establish the correctness of agent-produced code.
Work across frontend, backend, infrastructure, DevOps, and data topics as required by the problem.
Continuously evolve personal and team development workflows as models and agent capabilities advance.
Work with OEM customers to agree on requirements and develop plans for delivering solutions using Agile processes.
Ensure adherence to software quality standards throughout the development lifecycle.