Intel

AI Frameworks Engineer - Intern

68.9k PLN68 880 PLN/ rok./ rok.StażUmowa o staż

InternHybryda·GdańskPart-time·Umowa o staż

#453752·Dodano 16 dni temu·19

Źródło: Intel

Aplikuj teraz

Tech Stack / Keywords

PythonvLLMTritonSYCLPyTorchCUDAOpenCLC C++

Firma i stanowisko

Intel is building the open software stack that makes large language models run fast on Intel GPUs. This internship is part-time, typically lasting 12-24 months, for current university students based in Poland. The intern will join an engineering team focused on developing and optimizing LLM inference frameworks.

Wymagania

Current enrollment in a BS or MS degree in Computer Science, Computer Engineering, AI, Mathematics, Physics, or related field; student status required throughout internship.
Proficiency in Python programming.
Understanding of deep learning and neural networks, including transformer architecture and LLM inference basics.
Familiarity with Linux user environments.
Strong analytical and problem-solving skills with eagerness to learn and teamwork ability.
Clear and concise English communication skills.

Preferred Qualifications:

Hands-on experience with PyTorch.
Exposure to GPU programming with tools such as Triton, CUDA, SYCL, or OpenCL.
Working knowledge of C or C++.
Experience in testing and debugging Python code, including unit testing.
Familiarity with vLLM or other LLM inference frameworks.
Contributions to open-source projects or a public code portfolio.
Interest or coursework in computer architecture, parallel programming, or numerical methods.

Obowiązki

Develop and optimize AI framework software for large language model inference on Intel GPUs under senior engineers' guidance.
Implement, port, and tune compute kernels such as Triton and SYCL for transformer components including attention, sampling, normalization, mixture-of-experts, and quantization.
Run benchmarks and profile inference workloads to identify and report performance bottlenecks.
Write and extend unit tests, correctness checks, and performance regression tests.
Investigate and debug issues from the Python framework down to GPU kernel behavior.
Collaborate with framework, compiler, and kernel engineers worldwide and participate in code review and open-source community discussions.

Benefity

Competitive total compensation package with a minimum annual salary of 68,880 PLN.
Hybrid work model allowing a split between onsite and remote work.

Elastyczne godziny

Inne informacje

Candidates must be current university students based in Poland at the start and throughout the internship. The internship lasts 12-24 months and is part-time alongside studies. All qualified applicants will receive consideration without discrimination based on diverse characteristics protected by law.

Intel

22 aktywne oferty

Zobacz wszystkie oferty

Aplikuj teraz