Low-Level Kernel & CUDA Engineer

A technology company in the AI infrastructure and enterprise intelligence space is developing a system-level platform that enables instant local vectorization for existing data systems — without pipelines, migrations, or new infrastructure.

The company’s technology treats embeddings as native system resources, managed below application-level abstractions and designed for direct consumption by compute hardware. Its driver-level tensor retrieval approach creates a hardware-aware retrieval path that minimizes data copying, improves predictability, and removes bottlenecks in AI-driven data access.

This is an opportunity to join a highly technical engineering environment and work on low-level, performance-critical software for advanced AI platforms, including GPU-accelerated infrastructure, memory management, CUDA, and large-scale model architectures such as Transformers and MoE.

Responsibilities

• Develop low-level software components, drivers, and CUDA-based modules for advanced AI platforms.
• Write high-performance C++ and CUDA code with a strong focus on memory management, optimization, and runtime efficiency.
• Work on performance-critical infrastructure supporting large-scale Transformer and MoE architectures.
• Optimize system performance across GPU, memory, and compute-intensive workloads.
• Analyze bottlenecks, debug complex low-level issues, and improve reliability and scalability.
• Collaborate with AI, infrastructure, and platform engineering teams to support advanced model execution and deployment.
• Use modern AI tools to improve development, debugging, productivity, and engineering workflows.

Required Qualifications

• B.Sc. in Computer Science, Computer Engineering, or a related field – must.
• At least 3 years of experience in C++ development – must.
• Hands-on experience with CUDA – must.
• Strong low-level software development skills and ability to work on performance-critical systems.
• Strong analytical and problem-solving skills.
• Ability to work in a fast-moving startup environment and take ownership of complex technical challenges.

JOB NO. 500238