Remote AI Systems Research Engineer Intern

Yotta Labs; United States; On-site; USD 34,000 - 62,000; Part time

Benefits offered by this job

Competitive internship compensation

Flexible remote work environment

Fast path to full-time offer

Job summary

Yotta Labs is seeking a motivated Research Engineer Intern to work on Trainium, GPU kernels, and LLM systems optimization. Over a 12–16 week remote internship, you will own a well-scoped project at the intersection of AI Systems, Compiler and Runtime Optimization, Distributed Training & Inference, and Large Language Model Infrastructure.

You will implement and optimize kernels for Attention, GEMM, MoE, and quantization; build custom operators with CUDA, Triton, ROCm/HIP, or Neuron; profile and

Qualifications

• Pursuing BS, MS, or PhD in Computer Science, Computer Engineering, or related field.
• Strong Python and C++ programming skills.
• Solid understanding of GPU/accelerator architecture (memory, parallelism, occupancy).
• Experience writing CUDA, Triton, ROCm/HIP, or Neuron kernels.
• Familiarity with AI frameworks (PyTorch, Dynamo, LMCache) and profiling tools.
• Ability to work independently in a collaborative, remote environment.

Responsibilities

• Implement and optimize compute kernels for Attention, GEMM, MoE, and quantization on NVIDIA, AMD, or AWS Trainium.
• Build custom operators using CUDA, Triton, ROCm/HIP, or the Neuron SDK with PyTorch/XLA.
• Profile and improve inference performance in vLLM, SGLang, and our custom runtimes — kernel fusion, scheduling, KV-cache and memory optimizations.
• Build benchmarks, chase down performance regressions, and turn profiler traces into concrete speedups.
• Ship code upstream to open-source AI infrastructure projects, with tests and documentation.

Skills

Python

C++

GPU/accelerator architecture

PyTorch

Profiling tools

Education

BS/MS/PhD in CS/CE

Tools

CUDA

Triton

ROCm/HIP

Neuron SDK

PyTorch/XLA

Job description

Yotta Labs is seeking a motivated Research Engineer Intern to work on Trainium, GPU kernels, and LLM systems optimization. Over a 12–16 week remote internship, you will own a well-scoped project at the intersection of AI Systems, Compiler and Runtime Optimization, Distributed Training & Inference, and Large Language Model Infrastructure.

You will implement and optimize kernels for Attention, GEMM, MoE, and quantization; build custom operators with CUDA, Triton, ROCm/HIP, or Neuron; profile and