Generative AI & Edge Systems, Automotive (Principal/Architect)

PER International; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Relocation support

Job summary

PER International in San Jose, CA is partnering with a top-five semiconductor firm to hire Principal Engineer / Architect for Generative AI & Edge Systems in Automotive. The role focuses on developing GenAI and edge AI solutions for in-vehicle systems, collaborating across silicon, software, and automotive teams to deploy production-ready AI on constrained edge hardware.

You'll lead end-to-end optimization of LLMs for on-device deployment and drive research-to-production transitions in an agile,

Qualifications

• Master's degree in a related field and 10+ years software engineering experience, including GenAI/LLM systems.
• Hands-on with ML/DL frameworks and edge inference toolchains.
• Experience optimizing models for edge deployment with latency/memory considerations.

Responsibilities

• Research and develop GenAI and deep learning-based edge systems for in-vehicle domains (Infotainment, Speech, UI, multimodal ADAS).
• Lead integration of GenAI frameworks optimized for automotive edge compute and production deployment.
• Collaborate with silicon and software teams to define GenAI/LMM solutions with edge feasibility as a core constraint.
• Own end-to-end LLM optimization for on-device deployment on GPUs/NPUs, including quantization, pruning, and distillation.
• Track advances in edge AI and translate breakthroughs into deployable in-vehicle improvements.

Skills

GenAI/LLM systems

Edge AI optimization

C/C++/Python

Education

Master's degree in CS/EE/Math

Tools

TensorFlow

PyTorch

TensorRT

CUDA

ONNX Runtime

Job description

Position: Principal Engineer/Architect – Generative AI & Edge Systems, Automotive

Location: San Jose CA (4 days onsite)

Employment Type: Full-Time

Overview

We are partnering with a top-five semiconductor company to hire Principal Engineer / Architect - Generative AI & Edge Systems (Automotive).

This is an exciting opportunity to join a world-class automotive engineering team developing next-generation Generative AI and deep learning solutions for edge computing platforms. Working across silicon, software, and automotive customer teams, you will help bring cutting-edge AI technologies into production for future in-vehicle systems.

Key Responsibilities

• Research and develop Generative AI and deep learning-based systems for in-vehicle domains - Infotainment, Speech, UI, and multimodal ADAS - with a focus on architectures deployable on resource-constrained edge hardware.
• Lead integration of GenAI frameworks and toolchains optimized for automotive edge compute, driving adoption of cutting-edge on-device AI techniques from research into production.
• Partner with Silicon and Software engineering teams and automotive customers to define and deliver technical solutions for GenAI and LLM requirements, with edge feasibility as a core design constraint from the start.
• Own end-to-end LLM optimization for on-device deployment on GPUs and NPUs - profiling models, identifying bottlenecks, and driving improvements at both the model level (quantization, distillation, pruning) and framework level to maximize inference performance within embedded compute and power budgets.
• Track industry and academic advances in edge AI and efficient LLM techniques, translating relevant breakthroughs into practical, deployable improvements for in-vehicle systems.

Required Skills & Experience

• Master's degree in Computer Science, Electrical Engineering, Mathematics, or related field, with 10+ years of overall software engineering experience, including 4+ years directly relevant to GenAI/LLM systems.
• Deep hands-on experience with ML/DL frameworks and tooling: TensorFlow, PyTorch, NeMo, TAO, TensorRT, CUDA, and vLLM, with working knowledge of edge-inference frameworks (e.g., TensorRT-LLM, ONNX Runtime, GGUF/llama.cpp).
• Proven experience optimizing state-of-the-art models for edge deployment - quantization, pruning, distillation - with practical tradeoff analysis across latency, memory bandwidth, and CPU vs. NPU vs. GPU compute.
• Hands-on experience deploying and optimizing multiple LLMs and multimodal models concurrently on edge devices - managing shared compute/memory budgets, model orchestration, and runtime tradeoffs when running several models side-by-side on constrained hardware.
• Practical experience with speech and multimodal AI pipelines - ASR, TTS, speech-to-intent, or vision-language fusion - in resource-constrained, real-time environments.
• Strong programming proficiency in C, C++, and Python, including performance-critical and memory-constrained code.
• Hands-on experience with embedded software, RTOS, and microcontroller-based platforms.
• Familiarity with training data preparation, curation, and open-source datasets for fine-tuning or evaluation.
• Working knowledge of in-vehicle AI technical stacks - Speech, Voice, or ADAS/AD systems.
• Exposure to automotive functional safety (ISO 26262) and cybersecurity (ISO 21434) standards.
• Demonstrated ability to move quickly in agile environments, translating research and prototypes into shippable, production-grade systems.

Additional Information

• This is a hands-on Individual Contributor role at the Principal Engineer / Architect level.
• Candidates should be willing to work onsite in San Jose, CA.
• Relocation support is available for this role