Junior GPU kernel engineers in the US with CUDA/Triton experience

completed5 qualified1 runApr 21, 6:25 PMjunior-gpu-kernel-engineers-in-the-us-with-cudatriton-experi-1776795924

Parsed3 topics · Junior · Engineer · United States

Generating seed nodes

0 proposed

Explored 0 queries

0/0 done

Expanding nodes

queued

Qualifying candidates

queued

Qualified Candidates (5)

high hireability

PhD student@Graduate Student, Carnegie Mellon University

Previously: Deep Learning Architecture Intern @ NVIDIA

San Francisco, US

ML systems PhD (CMU Parallel Data Laboratory) with deep GPU expertise via pipeline parallelism and LLM training efficiency work (PipeFill at MLSys 2025, GraphPipe, Sia)
Work is GPU systems-level (scheduling, pipeline bubble utilization) rather than CUDA/Triton kernel engineering specifically — adjacent but not a direct match
Located in SF per DB
Hireability: HIGH — PhD completed ~2025 per DB records, likely on the job market as a recent graduate

high hireability

Research Assistant@PhD Student, Harvard University

Previously: Visiting Researcher @ Meta

Boston, US

ML systems PhD at Harvard (Stratos Idreos lab) with strong GPU systems work — co-authored Flash Inference (ICLR 2025) using kernel-level tiling to reduce memory movement (110× speedup in position-mixing), TorchTitan distributed training with Float8/SymmetricMemory kernel features, and SimpleFSDP via torch.compile
Work is at the PyTorch framework/compiler layer rather than direct CUDA/Triton kernel authorship, but demonstrates deep GPU memory hierarchy understanding
Based in Boston, US
Hireability: HIGH — PhD student at Harvard (pipeline-assessed Feb 2026 as final-year, expected May 2025 graduation), forked Harvard dissertation template, likely recently graduated and entering job market

medium hireability

CS PhD Student@Snowflake AI Research

Previously: Research Scientist Intern @ Snowflake

ML systems PhD at CMU with strong GPU kernel relevance: first-author on Korch (kernel orchestration for tensor programs, ASPLOS 2024) and core contributor to FlexFlow (C++/CUDA distributed DNN training)
Research focuses on parallel computing and LLM inference acceleration
US-based (Pittsburgh, PA)
Hireability: MEDIUM — 4th year PhD (expected graduation 2027), ~1-1.5 years from finishing; currently interning at Snowflake AI Research, indicating industry openness, but not yet in the prime final-year transition window

medium hireability

Ph.D. student@Carnegie Mellon University

Previously: MS student @ Carnegie Mellon University

Pittsburgh, US

PhD student at CMU Catalyst (Zhihao Jia lab) focused on GPU-accelerated LLM serving
Co-authored OSDI 2026 'Mirage Persistent Kernel' (mega-kernelizing tensor programs via compiler+runtime) and ASPLOS 2024 SpecInfer
GitHub has flashinfer (CUDA kernel lib for LLM serving) pinned
Strong CUDA/GPU kernel background, US-based in Pittsburgh
Hireability: MEDIUM-HIGH — ~5th year PhD based on 2021 first paper, likely approaching graduation; LinkedIn profile went fully empty on Jan 2026 scrape (possible deactivation or transition signal); website lists 'open to collaboration'

medium hireability

Ph.D. student@Carnegie Mellon University

Previously: Research Intern @ Meta

New York, US

Strong ML systems researcher at CMU (Beidi Chen / Zhihao Jia lab) focused on efficient LLM serving — SpecInfer, Sequoia, MagicDec, MagicPIG (LSH-based custom attention kernels), Mini-Sequence Transformer (memory-efficient CUDA training ops)
GPU kernel work evidenced by flashinfer (CUDA) and flash-linear-attention (Triton) forks, pytorch extension-cpp
H-index 9, multiple top-venue papers
In New York, US
Hireability: MEDIUM — 3rd year PhD (started 2023, likely 2 more years to graduation), but CV update 67 days ago + Meta FAIR 2025 internship shows active career motion; not yet in final-year transition window

#1completed0 qualified / 0 foundApr 21, 6:25 PM