Research Engineer Intern - AI Systems

Yotta
Regional
RemoteCareer-pivot friendly

Who this role is best for

Geared toward junior-level researchers comfortable with GPU kernel optimization and open-source contributions, this remote internship emphasizes hands-on technical work and collaboration across global teams.

Best fit for

  • Junior researchers with experience in AI framework optimization and open-source project contributions
    — “Strong interns receive return offers for full-time roles”
  • Candidates with a background in GPU/accelerator architecture and a drive to work on heterogeneous AI systems
    — “building the next generation multi-silicon AI cloud and runtime platform”
  • Individuals who can independently execute complex projects and deliver production-ready code
    — “taking it from design to working, profiled code running on real hardware”

Things to consider

  • The role demands a high level of self-direction and initiative despite being a remote internship
    — “the ability to work independently in a collaborative, remote environment”
  • A significant portion of the role involves direct code contributions to open-source projects
    — “Ship code upstream to open-source AI infrastructure projects, with tests and documentation”

How to stand out

  • Highlight personal projects or coursework in GPU kernel development and AI framework optimization
    — “Experience writing CUDA, Triton, ROCm/HIP, or Neuron kernels — class projects and personal projects count”
  • Emphasize any prior contributions to open-source AI infrastructure projects like vLLM or Triton
    — “Contributions to open-source AI infra projects like vLLM, SGLang, PyTorch, or Triton”
  • Demonstrate familiarity with profiling tools and performance debugging on real workloads
    — “Experience with profiling tools ... and performance debugging on real workloads”
  • Showcase knowledge of LLM inference internals such as FlashAttention or speculative decoding
    — “Familiarity with LLM inference internals — FlashAttention, PagedAttention, continuous batching, speculative decoding, MoE, or quantization”
Pace · Fast PacedCollaboration · HighAutonomy · MediumDecision Impact · TeamLevel · Intern

Derived from job-description analysis by Serendipath's career intelligence engine.

What success looks like

  • Implement and optimize compute kernels for Attention, GEMM, MoE, and quantization
  • Build custom operators using CUDA, Triton, ROCm/HIP, or the Neuron SDK
  • Profile and improve inference performance in vLLM, SGLang, and custom runtimes
Typical background
BS, MS, or PhD in Computer Science, Computer Engineering, or related field

Skills & requirements

Required

PythonCUDAPyTorchC++GPU ArchitectureDistributed Training & InferenceLarge Language Model Infrastructure

Preferred

TritonRocm/hipNeuron SDKLLM Inference Internals

Stack & domain

PythonCUDATritonRocm/hipNeuron SDKPytorch/xlaNvidia GpusAMD GpusAWS TrainiumVllmSglangLLM Inference InternalsFlashattentionPagedattentionContinuous BatchingSpeculative DecodingMoeQuantizationProfiling ToolsNsightRocm ProfilerNeuron ProfilerPytorch ProfilerProblem-solvingCommunicationCollaborationAI SystemsCompiler And Runtime OptimizationDistributed Training & InferenceGpu/accelerator Kernel DevelopmentLarge Language Model Infrastructure

About the role

Original posting from Yotta via Ashby

Location: Remote (Global)

Type: Internship

Company: Yotta Labs

Apply: careers@yottalabs.ai

🧠 About Yotta Labs

Yotta Labs is building the next generation multi-silicon AI cloud and runtime platform to power the world’s most demanding AI workloads. We enable training and inference across NVIDIA GPUs, AMD GPUs, and AWS Trainium, helping AI companies achieve the best performance and economics across heterogeneous hardware. Our mission is to provide high-performance AI computing and Model API services, enabling AI companies, research labs, and enterprises to train, deploy and integrate cutting-edge models at scale.

🛠️ Role Overview

We are seeking a highly motivated Research Engineer Intern to work on Trainium, GPU kernels, and LLM systems optimization. Over a 12–16 week internship, you will own a well-scoped project at the intersection of AI Systems, Compiler and Runtime Optimization, Distributed Training & Inference, GPU/Accelerator Kernel Development, and Large Language Model Infrastructure — taking it from design to working, profiled code running on real hardware. Your work will ship to production or open source and directly impact the performance of AI applications deployed on our platform. Strong interns receive return offers for full-time roles.

🎯 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.

✅ Qualifications

  • Currently pursuing a BS, MS, or PhD in Computer Science, Computer Engineering, or a related field.
  • Solid programming skills in Python and familiarity with C++.
  • Understanding of GPU/accelerator architecture fundamentals (memory hierarchy, parallelism, occupancy) from coursework, research, or projects.
  • Experience writing CUDA, Triton, ROCm/HIP, or Neuron kernels — class projects and personal projects count.
  • Strong understanding of AI frameworks (e.g., PyTorch, Dynamo, LMCache), model architectures and profiling tools (e.g. Nsight, ROCm Profiler, or Neuron Profiler).
  • Strong problem-solving skills and the ability to work independently in a collaborative, remote environment.

🌟 Preferred Experience

  • Contributions to open-source AI infra projects like vLLM, SGLang, PyTorch, or Triton.
  • Familiarity with LLM inference internals — FlashAttention, PagedAttention, continuous batching, speculative decoding, MoE, or quantization.
  • Experience with profiling tools (e.g. Nsight, ROCm Profiler, Neuron Profiler, or PyTorch Profiler) and performance debugging on real workloads.
  • Publications in top-tier conferences like MLSys, OSDI, SOSP, NSDI, SC, HPCA, or ISCA

🌐 Why Join Yotta Labs?

  • Be part of a visionary team aiming to redefine AI infrastructure and influence the future of multi-silicon AI computing.
  • Work on frontier AI infrastructure problems with access to serious hardware — latest-generation NVIDIA GPUs, AMD accelerators, and AWS Trainium at scale.
  • Get direct mentorship from engineers from leading institutions and tech companies.
  • Competitive internship compensation, a flexible remote work environment, and a fast path to a full-time return offer for top performers.

📩 How to Apply

Interested candidates should apply directly or send their resume to careers@yottalabs.ai. Please include links to any relevant projects or contributions (GitHub, open-source PRs, course projects) — for internships, these matter more to us than a cover letter.

Source: Yotta careers (Ashby)

Similar roles