Artificial Intelligence Researcher

A global high-tech company operates R&D centers in Israel, serving as an Innovation Hub for a global tech giant. These centers focus on end-to-end research and development, solving complex challenges across various domains, including communication, cybersecurity, and AI, working at the forefront of technology.

The company is based in Hod Hasharon, employs around 500 people in Israel, and offers a hybrid work model with two remote days per week.

Responsibilities-

• Join a team specializing in AI infrastructure and neural network architectures for LLMs.
• Work on both theoretical and engineering tasks in a hardware environment (no prior hardware experience required).
• Conduct research, implementation, and optimization of advanced deployment solutions for LLM models.
• Explore new directions for innovation and apply optimization techniques to Transformer-based models.
• Research and develop infrastructure components using PyTorch.
• Implement and optimize Model Serving solutions using vLLM, TGI, and MindIE.
• Collaborate with various teams to enhance system architecture and performance.
• Tech Stack: Python, C, C++, PyTorch, TensorFlow, TVM, TensorRT, CUDA, Attention mechanisms, LLM Serving frameworks, Optimization techniques.

Requirements:

• BSc in Computer Engineering or Computer Science.
• 5+ years of experience.
• Proficiency in Python and C++/C programming.
• Extensive experience with PyTorch and other deep learning frameworks.
• Deep understanding of Transformer architecture (Attention, MLP, KV cache).
• Experience with LLM serving frameworks and optimization techniques.
• Strong system-level understanding of hardware accelerators.
• Experience in system design and architecture.
• Proven experience in building and optimizing infrastructure components.
• Expertise in LLM deployment optimization, including Parallelization strategies, Scheduling optimization, and Batching strategies.
• Knowledge of memory optimization for large-scale models.