LLM Systems Engineer: Inference Hardware & Research

Netpreme; Cambridge (MA); On-site; USD 190,000 - 230,000; Full time

Benefits offered by this job

Relocation assistance

Visa sponsorship

Daily lunch stipend

401k match

Equity grant

Job summary

Netpreme is seeking a motivated LLM Systems Engineer to explore and prototype inference systems for next‑gen hardware. You’ll guide the hardware team on product definitions and push ML systems research forward in a hands‑on role.

The role blends engineering and research, requiring a track record in ML systems work and strong familiarity with modern LLM inference stacks. On-site roles available in Santa Clara, CA or Boston, MA.

Qualifications

• MS or PhD in computer systems with focus on ML inference or distributed systems.
• Experience contributing to core LLM inference infrastructures (vLLM, SGLang, TensorRT).
• Accelerator programming experience (CUDA, JAX/Pallas, ROCm).

Responsibilities

• Prototype and optimize emerging ML inference systems.
• Develop memory models for expandable vRAM.
• Write efficient GPU kernels for data movement.
• Benchmark inference engines in simulations and real hardware.

Skills

LLM inference systems

GPU kernel optimization

Memory modeling

Performance benchmarking

Design-space exploration

Education

MS/PhD in computer systems

Tools

CUDA

JAX/Pallas

ROCm

TensorRT

vLLM

SGLang

Job description

Netpreme is seeking a motivated LLM Systems Engineer to explore and prototype inference systems for next‑gen hardware. You’ll guide the hardware team on product definitions and push ML systems research forward in a hands‑on role.

The role blends engineering and research, requiring a track record in ML systems work and strong familiarity with modern LLM inference stacks. On-site roles available in Santa Clara, CA or Boston, MA.