Edge AI Model Optimization Software Engineer at NXP… · Guadalaja…

We at NXP have an environment that fosters innovation. Our team has technology experts who understand the big picture and mentors who coach passionate professionals to work on the most exciting challenges. We share responsibilities in everything we do, where every point of view is valued. Join us!

Job Summary

We are looking for a hands-on AI Optimization Engineer who thrives at the intersection of algorithms, systems engineering, and production-grade software. We are looking for someone who wants to work at the heart of state-of-the-art AI, not yesterday’s models, but the frontier: Generative AI, Transformers, Vision-Language Models (VLMs), and the emerging wave of Agentic AI systems.

In this role, you will contribute to the engineering of the core optimization technology that makes these advanced models run efficiently on NXP’s next-generation edge platforms, such as Ara 2. While Neural Network Quantization will be your primary focus, your impact will go far beyond it. You will architect high-performance software and collaborate across compiler, and hardware teams. Your work will directly define how frontier AI models are executed on-device, enabling fast, reliable, and power-efficient GenAI on resource-constrained platforms.

If you want to work where AI innovation meets real-world engineering, and build the infrastructure that defines what tomorrow’s smart, autonomous, AI-powered devices can do, this is the role for you.

Job Responsibilities

1. Optimization Tools: Own and evolve our production-grade optimization tools. You will design and implement quantization features, including mixed-precision flows that are robust enough for global deployment.

2. Pipelines: Design and maintain scalable PTQ (Post-Training Quantization) and QAT (Quantization-Aware Training) workflows, ensuring seamless integration with other optimizations and downstream deployment stacks.

3. Applied Innovation & Tooling: Collaborate on Proof‑of‑Concepts (POCs) for state-of-the-art quantization techniques. You won’t just prototype; you will evaluate the real-world impact of novel ideas and engineer the "bridge" that turns successful experiments into production-ready deployment recipes and developer tooling.

4. Numerical & Hardware Rigor: Apply your math foundation to implement approximation algorithms (range estimation, bias correction, BN-folding) while ensuring bit-exactness on target hardware. You will bridge the gap between abstract math and physical constraints like accumulator widths, saturation, and rounding behaviors.

5. Systems Performance: Profile and optimize the "hot paths" of our optimization toolchain to meet strict memory and compute-constrained targets.

6. Cross-Functional Leadership: Act as the technical bridge between AI Research and Hardware Engineering, providing quantified guidance on how choices impact model accuracy and performance.

7. Deployment Architecture: Document algorithmic tradeoffs and derive "gold-standard" deployment recipes, acting as technical mentor for other engineering teams, ensuring deployment strategies are scalable and repeatable.

Job Qualifications

Required Background

· Education: MSc or Ph.D. (is a plus) in Computer Science, Electrical Engineering, or Mathematics with a specialization in Machine Learning or Deep Learning.

· Systems Programming: Mastery of Python and C/C++. You should be comfortable with memory management and understanding how code maps to hardware (CPUs/NPUs).

· AI Expertise: Proven experience in AI/ML with a good understanding of CNN architectures and Generative AI (Transformers).

· AI Optimization: Experience with (or a strong desire to learn) quantization workflows and troubleshooting accuracy regressions.

· Technical Stack: Strong hands-on experience with PyTorch, ONNX, and other AI/ML frameworks.

Preferred

· Hardware Acceleration: Experience with hardware accelerators, device-level profiling, and diagnosing memory bottlenecks.

· Embedded Mindset: Familiarity with the constraints of embedded systems (latency, power, memory bandwidth).

· Advanced AI: Experience implementing state-of-the-art quantization for generative AI (e.g., GPTQ, Smoothquant, etc).

· Compilers: Knowledge of MLIR or TVM is a significant plus.

What You Will Gain

· Be part of a pioneering team shaping the future of AI and edge computing.

· Work on innovative projects that solve real-world challenges.

· Opportunity to grow with a dynamic, forward-thinking company.

· Competitive salary, benefits, and a collaborative work environment.

#LI-FCC3

More information about NXP in Mexico...

#LI-fcc3