Senior AI/ML Engineer: LLMs, On-Prem & Edge

Lattice Semiconductor Corp.; San Jose (CA); Hybrid; USD 188,000 - 254,000; Full time

Job summary

Lattice Semiconductor Corp. in San Jose is seeking a Sr. Staff Engineer to advance ML model development and deployment, focusing on LLMs and edge/on-prem solutions. You will architect end-to-end pipelines, optimize runtimes, and collaborate with cross-functional teams to deliver AI-powered hardware and software solutions.

The role emphasizes hands-on model development, mentoring engineers, and staying current with AI research while ensuring security and compliance in restricted environments.

Qualifications

• Master’s or Ph.D. in computer science, engineering, or related field, or equivalent experience.
• 8+ years in AI/ML with 3+ years on LLMs, code generation or large-scale NN.
• Hands-on experience with LLMs (GPT-OSS, Nemotron, Kimi, Qwen, LLaMA, Mistral, Falcon).
• Proficiency in Python and ML frameworks such as PyTorch or TensorFlow.

Responsibilities

• Train, fine-tune, and evaluate ML models including LLMs.
• Optimize models for local inference via quantization, pruning, distillation.
• Deploy models on-prem or at edge using PyTorch, TensorRT, ONNX, vLLM, llama.cpp.
• Build and maintain training and inference pipelines for reproducibility and scalability.
• Collaborate with software, infra teams, and domain experts to deliver AI solutions.

Skills

LLMs experience

Python

PyTorch

TensorFlow

Linux

Docker

Vector databases

Education

Master's or PhD in CS/Engineering

Tools

TensorRT

ONNX

llama.cpp

vLLM

LangGraph

Job description

Lattice Semiconductor Corp. in San Jose is seeking a Sr. Staff Engineer to advance ML model development and deployment, focusing on LLMs and edge/on-prem solutions. You will architect end-to-end pipelines, optimize runtimes, and collaborate with cross-functional teams to deliver AI-powered hardware and software solutions.

The role emphasizes hands-on model development, mentoring engineers, and staying current with AI research while ensuring security and compliance in restricted environments.