PhD Intern - Continuum Computing at Pacific Northwest National Laboratory

Wayne State University; Richland (WA); On-site; USD 34,000 - 55,000; Full time

Job summary

A research institution is seeking a PhD intern for Spring 2026 with a focus on High Performance Computing and Inference of Large Language Models. The candidate will engage in designing efficient cache management and developing strategies for LLM inference. Preferred qualifications include experience with open-source inference engines and GPU profilers. This internship offers flexibility in duration with a minimum of 3 months and can be remote or onsite.

Qualifications

• Candidate must have a strong background in High Performance Computing.
• Research experience demonstrated is preferred.
• Familiarity with Fabric Attached Memory protocols is a plus.
• Knowledge of CXL protocols and related hardware.

Responsibilities

• Design efficient KV cache management for distributed LLM Inference.
• Develop hybrid parallelism strategies for LLM inference.
• Participate in the development and publication of research.

Skills

High Performance Computing

Distributed LLM inference

Open-source inference engines

GPU profilers

CXL protocols

GPU profiling

CXL protocols

Education

Currently enrolled in a PhD program

Minimum GPA of 3.0

Job description

At PNNL, our core capabilities are divided among major departments that we refer to as Directorates within the Lab, focused on a specific area of scientific research or other function, with its own leadership team and dedicated budget.

Our Science & Technology directorates include National Security, Earth and Biological Sciences, Physical and Computational Sciences, and Energy and Environment. In addition, we have an Environmental Molecular Sciences Laboratory, a Department of Energy, Office of Science user facility housed on the PNNL campus.

The Physical and Computational Sciences Directorate's (PCSD’s) strengths in experimental, computational, and theoretical chemistry and materials science, together with our advanced computing, applied mathematics and data science capabilities, are central to the discovery mission we embrace at PNNL. But our most important resource is our people—experts across the range of scientific disciplines who team together to take on the biggest scientific challenges of our time.

The Advanced Computing, Mathematics, and Data Division (ACMDD) focuses on basic and applied computing research encompassing artificial intelligence, applied mathematics, computing technologies, and data and computational engineering. Our scientists and engineers apply end-to-end co-design principles to advance future energy-efficient computing systems and design the next generation of algorithms to analyze, model, understand, and control the behavior of complex systems in science, energy, and national security.

Responsibilities

The Future Technology Computing group seeks a PhD intern for Spring 2026 with a strong background on High Performance Computing and Inference of Large Language Models (LLMs). Knowledge of protocols and technologies supporting Fabric Attached Memory (FAM) is preferred but not strictly required. The internship can be either remote or onsite based on the availability of the candidate. The candidate will be expected to use and familiarize themselves with world leading hardware design tools which are available at the Pacific Northwest National Laboratory. The expected outcome involves high quality research work, represented by peer‑reviewed publications.

• Design efficient KV cache management for distributed LLM Inference on FAM‑based platforms.
• Develop hybrid parallelism strategies for LLM inference using low‑level programming models and runtimes.
• Participate in the develop and publication of a peer‑reviewed publication to present the proposed techniques.

The duration of the internship is flexible with a minimum of 3 months.

Qualifications

Minimum Qualifications:

• Candidates must be currently enrolled/matriculated in a PhD program at an accredited college.
• Minimum GPA of 3.0 is required.

Preferred Qualifications:

• Demonstrated research experience.
• Experience with distributed LLM inference.
• Experience working with open‑source inference engines for distributed serving.
• Knowledge of GPU profilers and performance analysis methods (including FLOPs/byte and data‑movement analysis).
• Experience with CXL protocols and related hardware.