Intern ML Compiler for Auto-Tuning (PyTorch/Triton)

samsungsemiconductor

San Jose (CA)

On-site

USD 55,000 - 83,000

Full time

5 days ago
Be an early applicant

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Benefits offered by this job

Medical/Dental/Vision
Inclusive rewards plan
Investment in your future
Relocation stipend
Housing stipend

Job summary

Samsung Semiconductor's ML Compiler team in San Jose seeks a driven graduate student to help build a cost-model-guided auto-tuning framework for PyTorch and Triton kernels on our compute-near-memory AI accelerator. You will connect it to torch.compile, measure gains on LLM inference, and collaborate with PyTorch and Triton open-source communities.

Location: daily onsite at our San Jose Headquarters with flexible work alignment.

Qualifications

  • Pursuing a Masters or PhD in Computer Science, Computer Engineering or Electrical Engineering; PhD preferred.
  • Must have at least 1 academic quarter/semester remaining.
  • Strong Python and C++, with experience writing or tuning Triton kernels or working with PyTorch compiler internals (torch.compile, FX, Inductor).
  • Grounding in compilers or performance engineering: IR and passes, cost models, roofline analysis and profiling.
  • Preferred: auto-tuning or auto-sharding experience (TVM/Ansor, Halide autoscheduler, Kernel Tuner, Alpa, GSPMD) or search methods like Bayesian optimisation or ILP.
  • Preferred: MLIR/LLVM back ends, scratchpad or near-memory architectures, LLM inference kernels, or open-source PyTorch/Triton contributions.

Responsibilities

  • Design a search space for Triton kernel configurations on the accelerator: tile sizes, launch-grid mapping, on-chip buffer budgets and load style.
  • Build legality rules with our compiler engineers that prune configurations that will not compile or will not fit in on-chip memory.
  • Develop a kernel cost model, calibrated against our simulator and profiler, that ranks candidates so only the most promising are benchmarked.
  • Build a reliable benchmarking harness and a persistent tuning cache, so that tuning cost stays practical.
  • Integrate the tuner with the torch.compile Inductor autotuning path and with Triton's autotuner.
  • Extend tuning to data placement: tune kernels for a given tensor layout and export tuned costs that a placement pass can use.
  • Measure kernel-level and end-to-end LLM inference gains, and share results in weekly syncs and a final presentation.
  • Collaborate with PyTorch and Triton developers on upstream autotuning hooks, and hand off documented, tested code.
  • Complete other responsibilities as assigned.

Skills

Performance optimisation
PyTorch internals
LLM benchmarking

Education

Masters/PhD in CS/CE/EE

Tools

PyTorch
Triton
LLVM

Job description

Samsung Semiconductor's ML Compiler team in San Jose seeks a driven graduate student to help build a cost-model-guided auto-tuning framework for PyTorch and Triton kernels on our compute-near-memory AI accelerator. You will connect it to torch.compile, measure gains on LLM inference, and collaborate with PyTorch and Triton open-source communities.

Location: daily onsite at our San Jose Headquarters with flexible work alignment.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

ML Compiler Intern: Auto-Tuning for PyTorch & Triton Kernels
ML Compiler Intern: Auto-Tuning for PyTorch & Triton Kernels

Samsung Electronics America • San Jose (CA)

On-site
USD 39,000 - 94,000
Relocation Support
Housing Stipend
Give Back
+3
Intern, ML Compiler (Kernel Auto-Tuning)
Intern, ML Compiler (Kernel Auto-Tuning)

samsungsemiconductor • San Jose (CA)

On-site
USD 55,000 - 83,000
Medical/Dental/Vision
Inclusive rewards plan
Investment in your future
+2
Staff Compiler Engineer, PyTorch — Hybrid/Onsite IR & DSL
Staff Compiler Engineer, PyTorch — Hybrid/Onsite IR & DSL

Samsung Electronics America • San Jose (CA)

On-site
USD 163,000 - 253,000
4+ weeks paid time off
Diverse benefits package for wellbeing
Support for family care and emotional wellness
Intern, ML Compiler (Kernel Auto-Tuning)
Intern, ML Compiler (Kernel Auto-Tuning)

Samsung Electronics America • San Jose (CA)

On-site
USD 39,000 - 94,000
Relocation Support
Housing Stipend
Give Back
+3
Senior ML Compiler Architect
Senior ML Compiler Architect

Samsung Electronics America • San Jose (CA)

On-site
USD 163,000 - 253,000
4+ weeks PTO
Fertility/adoption stipend
Emotional wellness program
+2
Senior AI Compiler Engineer - Triton & MLIR Expert
Senior AI Compiler Engineer - Triton & MLIR Expert

NVIDIA Corporation • United States

Remote
USD 184,000 - 288,000
Equity eligibility
Benefits package
Systems ML Engineer: AI Compiler & Kernel Innovator
Systems ML Engineer: AI Compiler & Kernel Innovator

Meta • New York (NY)

On-site
USD 122,000 - 181,000
Bonus
Equity
Benefits
ML Compiler Intern: Shape the Future of AI/ML
ML Compiler Intern: Shape the Future of AI/ML

samsungsemiconductor • San Jose (CA)

On-site
USD 55,000 - 83,000
Relocation Support
Housing Stipend
Charitable giving match
Senior ML Compiler Engineer — Optimize Models for Accelerators
Senior ML Compiler Engineer — Optimize Models for Accelerators

Annapurna Labs • Cupertino (CA)

On-site
USD 193,000 - 262,000
Health insurance
401(k) matching
RSUs / Stock options
ML Compiler Intern: Optimize Deep Learning on Advanced Hardware
ML Compiler Intern: Optimize Deep Learning on Advanced Hardware

Samsung Electronics America • San Jose (CA)

On-site
USD 39,000 - 94,000
Relocation Support
Housing Stipend
Give Back
+3