Senior Agent Researcher / Engineer
Our client is a Singapore-based automotive tech company, building AI-powered computing platforms for software-defined vehicles. They specialize in intelligent cockpit systems and autonomous driving computation, combining custom automotive SoCs with advanced software to help automakers deliver safer, smarter, more personalized in-car experiences.
Senior Agent Researcher / Engineer
Role Overview:
As a Senior Agent Researcher, you will design and build evaluation-driven Agent and multi-agent systems for complex real-world applications. You will focus not just on making Agents work, but on understanding, measuring, and continuously improving their behavior, combining research depth with strong engineering execution in an AI-native development environment.
Key Responsibilities:
- Build and evolve evaluation-driven Agent systems, using evaluation as the core mechanism for behavior improvement and iteration
- Design Agent architectures that are observable, maintainable, and optimizable under real-world constraints
- Develop scalable evaluation infrastructure, including automated testing, benchmarking, and experimentation frameworks
- Optimize Agent performance acrossarchitecture, model selection, context management, memory, and tool usage,balancing performance and cost
- Rapidly prototype Agent systems and demos,integrating research insights with engineering and product validation
- Continuously explore and apply advancements in LLMs, multimodal models, multi-agent systems, and context engineering
Qualifications & Requirements:
- Proven experience designing, building, or deploying Agent systems, LLM-based applications, or complex AI systems in production environments
- Strong understanding of AI/ML fundamentals and system design, including models, inference, training, context, and evaluation
- Hands-on experience with Agent evaluation frameworks, automated experimentation, or A/B testing
- Ability to translate cutting-edge research(LLMs, MLLMs, multi-agent systems) into practical, production-ready solutions
- Strong systems thinking with a research-oriented mindset and solid engineering execution
- Publications, open-source contributions, or experience in automotive, chip, or intelligent cockpit systems are advantageous
EA Personnel Registration No: R1106329 EA License No: 12C6254