Principal LLM Inference Engineer — Agentic AI Systems

United States Digital Space LLC; United States; Remote; USD 272,000 - 431,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

United States Digital Space LLC is seeking outstanding engineers to advance LLM inference, improve performance, and develop scalable systems for agentic workflows. You will explore generative AI research and implement efficient inference algorithms within our software stack, collaborating across teams and partners to define system requirements.

Ideal candidates have 15+ years in deep learning systems, strong Python and C++ skills, and a solid grasp of computer architecture and GPU-based

Qualifications

• BS/MS/PhD in CS/EE/CE or related field (or equivalent experience).
• 15+ years of experience in deep learning and deep learning systems design.
• Proficiency in Python and C++ programming.
• Strong understanding of computer architecture, and GPU/parallel datacenter computing fundamentals.

Responsibilities

• Research and Development: Explore contemporary research on generative AI, agents, and inference systems into the LLM software stack.
• Workload Analysis and Optimization: Profile and optimize agentic LLM workloads to reduce latency and increase throughput.
• System Design and Implementation: Design scalable systems to accelerate agentic workflows and handle datacenter-scale use cases.
• Collaboration and Communication: Advise on software/hardware/system iterations by engaging with teams and external partners.

Skills

Deep learning

Python

C++

GPU computing

Performance modeling

Education

BS/MS/PhD in CS/EE/CE or related field

Tools

CUDA

OpenCL

Job description

United States Digital Space LLC is seeking outstanding engineers to advance LLM inference, improve performance, and develop scalable systems for agentic workflows. You will explore generative AI research and implement efficient inference algorithms within our software stack, collaborating across teams and partners to define system requirements.

Ideal candidates have 15+ years in deep learning systems, strong Python and C++ skills, and a solid grasp of computer architecture and GPU-based