ML Engineer II: Multi-Agent AI & Inference Optimizer

UST; Oregon (WI); On-site; USD 82,000 - 123,000; Full time

Benefits offered by this job

Paid vacation days

Paid sick leave

Paid holidays

401(k) with employer matching

Medical insurance

Dental insurance

Vision insurance

Life insurance

Disability insurance

HSA/FSA options

Job summary

UST in the United States seeks a Machine Learning Engineer II to develop multi-agent workloads, optimize inference across OMIX/compute stacks, and improve latency and throughput. The role emphasizes hands-on Python and C/C++, GitHub usage, and proficiency with frameworks like LangGraph/LangChain; location Oregon; compensation is $82,000-$123,000 per year with full benefits.

The engineer will work across model selection, evaluation, and deployment for diverse agent workloads, leveraging

Qualifications

• Hands-on development of multi-agent workloads.
• Experience with agent frameworks such as LangGraph/LangChain, AutoGen, CrewAI etc.
• Strong understanding of LLMs, SLMs and multimodal models, including Transformer architecture, attention, tokenization, context windows, and KV cache.
• Hands-on experience with model selection, evaluation, and deployment for different agentic workload requirements.

Responsibilities

• Develop a multi-agent workload at the application level and optimize its execution across the stack.
• Practical use of GitHub Copilot for development, debugging and code generation.
• Optimize for latency, throughput, tokens/sec, memory footprint, and accelerator utilization.
• Select and tune models/inference engines based on workload characteristics and hardware capabilities.

Skills

Multi-agent workloads

LangGraph/LangChain

LLMs/SLMs

Python

C/C++

Git/GitHub

Linux shell

Docker

Performance engineering

Profiling

GPU memory management

Tools

OpenVINO

ONNX Runtime

vLLM

llama.cpp

TGI

CUDA

Job description

UST in the United States seeks a Machine Learning Engineer II to develop multi-agent workloads, optimize inference across OMIX/compute stacks, and improve latency and throughput. The role emphasizes hands-on Python and C/C++, GitHub usage, and proficiency with frameworks like LangGraph/LangChain; location Oregon; compensation is $82,000-$123,000 per year with full benefits.

The engineer will work across model selection, evaluation, and deployment for diverse agent workloads, leveraging