Software Engineer - AI Inference for Science
Argonne National Laboratory2 months ago
Lemont, IL, USAMid Level / Senior
Base Salary
$94k - $147k/yr
Responsibilities
- Engineer AI inference solutions integrated with scientific workflows and HPC resources.
- Provide programmatic AI model access through APIs and process large batches of prompts in parallel.
- Deploy and adapt evolving AI models while managing compute, memory, latency, and accelerator utilization.
- Design or operate distributed inference and data services, including routing, asynchronous execution, queueing, fault tolerance, and monitoring.
- Integrate services with HPC schedulers and manage resource provisioning, job lifecycles, and mixed latency- and throughput-oriented workloads.
- Optimize inference through batching, memory management, model parallelism, quantization, and accelerator utilization.
- Collaborate with science application teams and contribute to broader scientific initiatives.
- Create, maintain, and support high-quality domain-specific software and models.
Requirements
- Bachelor’s degree and 5+ years of experience, Master’s degree and 3+ years of experience, or PhD for RD2; equivalent qualifications are accepted.
- Bachelor’s degree and 8+ years of experience, Master’s degree and 5+ years of experience, or PhD and 4+ years of experience for RD3; equivalent qualifications are accepted.
- Familiarity with AI model use, deployment, and memory and compute requirements.
- Familiarity with AI hosting frameworks such as vLLM or SGLang.
- Familiarity with Python REST API development using frameworks such as FastAPI.
- Comprehensive programming experience in Python, C, or C++.
- Experience with version control software such as git.
- Preferred experience with distributed inference or data services, HPC schedulers such as Slurm or PBS, AI inference optimization, secure multi-user services, authentication and authorization, API security, and regulated environments.
- Preferred familiarity with Prometheus, Grafana, log aggregators, OLAP, or data warehousing techniques.
- Ability to work collaboratively and communicate effectively in writing and orally.
Benefits
- Hybrid remote work with employees typically working more than 60% of their time remotely and some days onsite.
- Regular, full-time employment with comprehensive employee benefits.
- Employment is contingent on a background check, and the position may require government access authorization after hire.
- Employees and certain guest researchers or contractors must disclose participation in Foreign Government Sponsored or Affiliated Activities for review.
Categories
About Argonne National Laboratory
Argonne National Laboratory is a U.S. Department of Energy Office of Science laboratory that conducts research in energy, climate, materials, computing, and national security. Headquartered in Lemont, Illinois and founded in 1946, it is managed by UChicago Argonne, LLC and operates major user facilities such as the Advanced Photon Source and the Argonne Leadership Computing Facility. Work is funded by federal programs with partnerships and technology transfer to industry and academia.