Senior Member of Technical Staff (Model Routing)

About Company

Founded in 2022, XG Tech is driving the future of smart vehicles. Its mission is to empower the digital transformation of automobiles, moving from distributed computing to a centralized, cross-domain platform.

XG Tech focuses on the intelligent cockpit—the next frontier of differentiation—while seamlessly integrating advanced driving systems. By reimagining cars as mobile living spaces, XG Tech aligns with the evolving trend of vehicles becoming the “third living space.”

Role Summary

As a Research Engineer, you will research, design, and develop intelligent model routing systems for personal AI computers, determining whether tasks should be processed by on-device small language models or escalated to frontier cloud models. You will develop adaptive routing strategies to optimize model selection based on task complexity, response quality, inference latency, computational efficiency, and cloud costs, ensuring high-quality AI experiences while maximizing local inference efficiency.

Key Responsibilities

• Design and train difficulty estimators and routers that choose between model tiers per request, per subtask, and per agent step — including mid-trajectory decisions where the right answer changes halfway through a task.
• Hold the quality line while cutting cost: our working target is a substantial reduction in cloud spend at unchanged task success on a fixed, agreed evaluation slice.
• Study the boundary of small-model capability — when the local model is genuinely out of its depth, how to detect that cheaply and early, and how to escalate without the user noticing a seam.
• Build the routing evaluation harness: cost accounting, latency measurement, and success metrics that reflect what the user experienced rather than what the router optimised.
• Evaluate routers honestly, including out-of-distribution behaviour and adversarial prompts; publish results and negative findings internally, and externally where they hold up.
• Work directly with the embedded and post-training teams so that routing decisions match what the device can actually run and what the model can actually do.

How you will stand out

• A strong academic or research background in Machine Learning, NLP, Computer Science, Systems, or a related field, with a PhD or equivalent practical research experience preferred
• Demonstrated experience in LLM research and development, particularly in model serving, evaluation, inference optimization, or intelligent model selection
• Strong understanding of model routing strategies and the trade-offs between response quality, inference latency, computational resources, and cloud costs
• Hands-on experience designing and conducting experiments, evaluating model performance, and developing data-driven approaches to optimize routing decisions
• Ability to independently identify research problems, formulate hypotheses, implement solutions, and analyze experimental results
• A strong track record of technical contributions demonstrated through research publications, production systems, open-source projects, or competitive achievements

Nice to have:

• Experience with speculative decoding, model quantization, model compression, or on-device inference
• Familiarity with AI agents and agent frameworks, particularly dynamic model routing during multi-step agent execution
• Experience with privacy-preserving inference or collaborative device–cloud AI architectures

XG Tech is committed to providing equal employment opportunities by country, state, and local laws. XG Tech does not discriminate against employees or applicants based on conditions such as race, color, gender identity and/or expression, sexual orientation, marital and/or parental status, religion, political opinion, nationality, ethnic background or social origin, social status, disability, age, indigenous status, and union.