Senior Deep Learning Software Engineer, Inference
From 18 400 PLN (UoP)
Wymagania
- C++
- Python
- CUDA
- PyTorch
- vLLM
Opis stanowiska
Senior Deep Learning Software Engineer, Inference
NVIDIA seeks a Senior Software Engineer specializing in Deep Learning Inference for our growing team. As a key contributor, you will help design, build, and optimize the GPU-accelerated software that powers today’s most sophisticated AI applications. Our team is responsible for developing and maintaining high-performance deep learning frameworks, including SGLang and vLLM, which are at the forefront of efficient large-scale model serving and inference. You will play a central role in improving these platforms, facilitating smooth deployment and serving of groundbreaking language models.
You’ll work closely with the deep learning community to implement the latest algorithms for public release in frameworks like SGLang and vLLM, as well as other DL frameworks. Your work will focus on identifying and driving performance improvements for state-of-the-art LLM and Generative AI models across NVIDIA accelerators, from datacenter GPUs to edge SoCs. You'll bring to bear open-source tools and plugins—including CUTLASS, OAI Triton, NCCL, and CUDA kernels—to implement and optimize model serving pipelines.
What you'll be doing:
- Performance optimization, analysis, and tuning of DL models in various domains like LLM, Multimodal and Generative AI.
- Scale performance of DL models across different architectures and types of NVIDIA accelerators.
- Contribute features and code to NVIDIA’s inference libraries, vLLM and SGLang, FlashInfer and LLM software solutions.
- Work with cross-collaborative teams across frameworks, NVIDIA libraries and inference optimization innovative solutions.
NVIDIA is at the forefront of breakthroughs in Artificial Intelligence, High-Performance Computing, and Visualization. Our teams are composed of driven, innovative professionals dedicated to pushing the boundaries of technology.
NVIDIA seeks a Senior Software Engineer specializing in Deep Learning Inference for our growing team. As a key contributor, you will help design, build, and optimize the GPU-accelerated software that powers today’s most sophisticated AI applications. Our team is responsible for developing and maintaining high-performance deep learning frameworks, including SGLang and vLLM, which are at the forefront of efficient large-scale model serving and inference. You will play a central role in improving these platforms, facilitating smooth deployment and serving of groundbreaking language models.
You’ll work closely with the deep learning community to implement the latest algorithms for public release in frameworks like SGLang and vLLM, as well as other DL frameworks. Your work will focus on identifying and driving performance improvements for state-of-the-art LLM and Generative AI models across NVIDIA accelerators, from datacenter GPUs to edge SoCs. You'll bring to bear open-source tools and plugins—including CUTLASS, OAI Triton, NCCL, and CUDA kernels—to implement and optimize model serving pipelines.
What you'll be doing:
- Performance optimization, analysis, and tuning of DL models in various domains like LLM, Multimodal and Generative AI.
- Scale performance of DL models across different architectures and types of NVIDIA accelerators.
- Contribute features and code to NVIDIA’s inference libraries, vLLM and SGLang, FlashInfer and LLM software solutions.
- Work with cross-collaborative teams across frameworks, NVIDIA libraries and inference optimization innovative solutions.
NVIDIA is at the forefront of breakthroughs in Artificial Intelligence, High-Performance Computing, and Visualization. Our teams are composed of driven, innovative professionals dedicated to pushing the boundaries of technology.
🔍 Dekoder Ogłoszenia
🔴
key contributor
Oczekuje się, że będziesz samodzielnie realizować zadania i wnosić znaczący wkład w projekty, a nie tylko wykonywać polecenia.
🟡
at the forefront of efficient large-scale model serving and inference
Praca może polegać na utrzymaniu i rozwijaniu istniejących, zaawansowanych technologii, a niekoniecznie na tworzeniu zupełnie nowych.
🟡
implement the latest algorithms for public release
Może to oznaczać implementację algorytmów zaproponowanych przez innych, a niekoniecznie tworzenie własnych innowacyjnych rozwiązań od podstaw.
🔴
identify and drive performance improvements
Oczekuje się proaktywnego wyszukiwania problemów z wydajnością i inicjowania działań naprawczych, co może być czasochłonne.
🟡
bring to bear open-source tools and plugins
Będziesz intensywnie korzystać z istniejących narzędzi open-source, co wymaga dobrej znajomości ekosystemu i potencjalnie rozwiązywania problemów związanych z integracją.