A Career at HARMAN

As a technology leader that is rapidly on the move, HARMAN is filled with people who are focused on making life better. Innovation, inclusivity and teamwork are a part of our DNA. When you add that to the challenges we take on and solve together, you’ll discover that at HARMAN you can grow, make a difference and be proud of the work you do every day.

A Career at HARMAN Automotive

We’re a global, multi-disciplinary team that’s putting the innovative power of technology to work and transforming tomorrow. At HARMAN Automotive, we give you the keys to fast-track your career

• Engineer audio systems and integrated technology platforms that augment the driving experience
• Combine ingenuity, in-depth research, and a spirit of collaboration with design and engineering excellence
• Advance in-vehicle infotainment, safety, efficiency, and enjoyment

What You’ll Be Part Of

As an LLM Application Engineer, you will be responsible for the design, implementation, iteration and optimization of multi-agent application systems. This is an individual-contributor role that values hands-on delivery experience across multi-agent paradigms, LLM reasoning orchestration, and agent tool-calling capabilities. You will work closely with software platform, product, algorithm and test teams to deliver stable and production-ready generative AI platform components. In this role, you will report to the AI Software Engineering Manager

How You’ll Make An Impact

• Lead the design and development of multi-agent orchestration modules, and build agent workflows for engineering-oriented scenarios
• Act as the key technical owner for Skill Agent design, ReAct reasoning paradigm implementation, prompt engineering, tool integration, agent memory management, workflow debugging and performance tuning
• Continuously partner with software platform, product management, algorithm and QA teams to meet system functional requirements, stability targets and non-functional metrics
• Participate in development and iteration of production-grade agent components, and deliver reusable capability modules for LLM-powered applications
• Drive the establishment of best practices for LLM-agent development, exception handling and runtime governance
• Conduct effect evaluation for agent systems, troubleshoot issues such as LLM hallucinations, tool-call failures and workflow dead-loops, and continuously improve agent performance
• Participate in security and risk governance for AI-powered applications, implementing constraints and validation rules for agent outputs and tool permissions
• Define interface specifications for agent components and produce technical documentation to enable capability reuse for other engineering team members

What Makes You A Strong Fit

• Solid theoretical understanding and hands-on practical experience with LLM multi-agent concepts, including Skill Agent, ReAct reasoning paradigm, Agent Harness and tool-calling mechanisms
• Proficient in Python programming, capable of independent coding, debugging and root-cause analysis for LLM-based applications
• Practical experience in prompt engineering, agent state management, RAG pipeline construction and troubleshooting
• Ability to build multi-agent workflows, integrate third-party tools / APIs into agent systems, and validate LLM-application functional and performance outcomes
• Understand inherent limitations of large language models; possess engineering-minded solutions to mitigate hallucinations, unstable outputs and context overflow problems
• Hands-on capability for AI application evaluation, able to design test cases and assess agent quality across functionality, stability and output fidelity dimensions
• Basic awareness of AI application security, including risk validation for user inputs and agent tool invocation permissions
• Ability to produce well-structured technical design documents and API specifications to support internal collaboration and capability reuse

Bonus Points if You Have

• Master’s degree in computer science, Artificial Intelligence or related technical disciplines
• Practical experience working with vector databases, and familiarity with containerized deployment practices for LLM services
• Good awareness of trade-offs among different agent paradigms, with hands-on experience adapting agent-oriented projects
• Hands-on experience building observability and evaluation frameworks for generative AI applications
• Project experience involving complex business workflow orchestration, task decomposition and task scheduling
• Practical knowledge of LLM context window management, hierarchical memory design and session-state persistence
• Proven track record optimizing AI system performance covering inference latency, throughput and resource consumption

Things To Know

• You have a working knowledge of English
• You are willing to travel, if necessary
• Due to the nature of global infrastructure, you understand that flexibility in working hours is required
• You can obtain a government security clearance; You are eligible to work in China

What You Can Expect

• Flexible work environment, allowing for full-time remote work one day per week for positions that can be performed outside a HARMAN or customer location

• Access to employee discounts on world-class Harman (JBL, HARMAN Kardon, AKG, etc.)

• Extensive training opportunities through our own HARMAN University

• Competitive wellness benefits

• Access to HARMAN Campus Fitness Center and Cafeteria

• “Be Brilliant” employee recognition and rewards program

• An inclusive and diverse work environment that fosters and encourages professional and personal development

HARMAN is proud to be an Equal Opportunity / Affirmative Action employer. All qualified applicants will receive consideration for employment without regard to race, religion, color, national origin, gender (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender identity, gender expression, age, status as a protected veteran, status as an individual with a disability, or other applicable legally protected characteristics.