AI Inference Engineer for Scientific HPC

Socket.dev; Lemont (IL); On-site; USD 94,000 - 147,000; Full time

Job summary

Argonne National Laboratory is seeking a Software Engineer to enable AI for science by building scalable inference within scientific workflows on HPC platforms and AI accelerators.

The role involves API-based access to AI models, parallel processing of prompts, and collaboration across science teams in a hybrid remote onsite setup. The successful candidate will work with Python, C/C++, and modern AI hosting tools.

Qualifications

• Candidate has familiarity with AI models, deployment, and memory/compute requirements.
• Experience with AI hosting frameworks like vLLM or SGLang is preferred.
• Proficiency in REST API development in Python using FastAPI.
• Strong programming skills in Python and C/C++, with version control (git).

Responsibilities

• Develop AI inference capabilities within scientific workflows.
• Engineer scalable solutions for HPC systems and AI accelerators.
• Collaborate with science teams and other labs to deploy models.
• Contribute to domain-specific software and model maintenance.

Skills

AI model usage

API development

Python

C/C++

Git

Collaborative work

Communication

AI accelerators

OpenAI API

Education

Bachelor’s degree in CS/CE or related

Tools

vLLM

SGLang

FastAPI

Job description

Argonne National Laboratory is seeking a Software Engineer to enable AI for science by building scalable inference within scientific workflows on HPC platforms and AI accelerators.

The role involves API-based access to AI models, parallel processing of prompts, and collaboration across science teams in a hybrid remote onsite setup. The successful candidate will work with Python, C/C++, and modern AI hosting tools.