10 дней назад
AI Frameworks Engineer - Intern
68 880PLN
Вакансия из Hirify Global, списка международных tech-компанийДля мэтча и отклика нужен Plus
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
AI Frameworks Engineer - Intern (AI/LLM inference): Developing and optimizing open-source inference frameworks and GPU kernels for large language models on Intel GPUs with an accent on transformer kernels, benchmarking, and correctness testing. Focus on implementing Triton and SYCL kernels, profiling performance bottlenecks, debugging across the software and hardware stack, and contributing to vLLM.
Location: Gdansk, Poland. Hybrid work model with time split between the assigned site and off-site work. Candidates must be based in Poland on the internship start date and throughout the internship.
Minimum annual base salary: 68,880 PLN.
Company
is building an open software stack for fast large language model inference on GPUs.
What you will do
- Develop and optimize AI framework software for LLM inference on GPUs, contributing upstream to open-source projects such as vLLM.
- Implement, port, and tune Triton and SYCL compute kernels for attention, sampling, normalization, mixture-of-experts, and quantization.
- Benchmark and profile inference workloads to identify and measure performance bottlenecks.
- Write unit tests, correctness checks, and performance regression tests for framework and kernel code.
- Investigate and debug issues across Python framework code, GPU kernels, and hardware behavior.
- Collaborate with framework, compiler, and kernel engineers through code reviews and technical discussions.
Requirements
- Current BS or MS student in Computer Science, Computer Engineering, AI, Mathematics, Physics, or a related field, with student status maintained throughout the internship.
- Good Python programming skills and an understanding of deep learning, neural networks, transformer architecture, and LLM inference.
- Familiarity with Linux environments and strong analytical and problem-solving skills.
- Must be based in Poland for the full duration of the internship.
- Clear and concise communication in English is required.
Nice to have
- Hands-on experience with PyTorch.
- Exposure to GPU programming with Triton, CUDA, SYCL, OpenCL, or similar technologies.
- Working knowledge of C or C++ and experience testing and debugging Python code.
- Familiarity with vLLM or another LLM inference serving framework.
- Open-source contributions, a public code portfolio, or coursework in computer architecture, parallel programming, or numerical methods.
Culture & Benefits
- Part-time internship alongside a university degree.
- Typical internship duration of 12–24 months.
- Mentorship and code review from experienced engineers.
- Contributions can reach open-source projects used across the AI industry.
- Collaboration with global engineering teams and the open-source community.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →