CUDA Developer

CUDA Developer

Location: LATAM

Experience Required: Minimum 5+ years of experience in software development with a focus on CUDA and C++. We are seeking experienced CUDA Developers to join our dynamic team. The ideal candidates will possess a strong background in parallel computing, algorithm optimization, and large-scale performance tuning using CUDA. This role is crucial for building and optimizing high-performance solutions, with a focus on leveraging cutting-edge technologies like CUDA and large language models.

Mandatory Skills

• CUDA: 2+ years
• C++: 5+ years
• Large Language Models (LLM): 2+ years

Responsibilities

• Identify and resolve performance issues and bugs related to CUDA implementations.
• Analyze and optimize algorithm performance for maximum efficiency through parallelization.
• Collaborate with cross-functional teams to integrate CUDA optimizations into existing software architecture.
• Stay updated with the latest trends, tools, and best practices in CUDA and parallel computing.
• Develop and maintain code for GPU-accelerated applications using CUDA.
• Implement code optimizations for memory usage and computational efficiency.

Required Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, or related fields.
• Minimum of 5 years of software development experience, with at least 2 years specifically in CUDA.
• Proven experience in optimizing performance and debugging CUDA applications.
• Ability to work both independently and collaboratively within a team.
• Strong problem-solving skills and attention to detail.
• Excellent communication skills to effectively work with technical and non-technical teams.

Knowledge and Skills

• Proficiency in C/C++ programming languages.
• Expertise in CUDA Performance Optimization and best practices.
• Experience with CUDA version 12.0 or above (preferably 12.3).
• Familiarity with frameworks and libraries such as cuBLAS, cuDNN, and Thrust.
• Strong understanding of GPU architectures and resource management.
• Knowledge of parallel computing patterns, memory management, and multi-threading.
• Experience with Large Language Models (LLMs) and their deployment on GPU infrastructure.

Preferred Qualifications

• Experience in developing or optimizing deep learning frameworks.
• Hands-on experience with CUDA Graphs or other advanced GPU programming techniques.
• Familiarity with AI/ML models and their training/inference processes.

About Us: InterSources Inc

InterSources Inc, a Certified Diverse Supplier, was founded in 2007 and offers innovative solutions to help clients with Digital Transformations across various domains and industries. Our history spans over 16 years and today we are an Award-Winning Global Software Consultancy solving complex problems with technology. We recognize that our employees and our clients are our strengths as the diverse talents and opportunities they bring to the table enable us to grow as a global platform and they are causally linked with our success. We provide strategic and technical advice, and we have expertise in areas covering Artificial Intelligence, Cloud Migration, Custom Software Development, Data Analytics Infrastructure & Cloud Solutions, Cyber Security Services, etc. We make reasonable accommodations for clients and employees and we do not discriminate based on any protected attribute including race, religion, color, national origin, gender sexual orientation, gender identity, age, or marital status. We also are a Google Cloud partner company. We align strategy with execution and provide secure service solutions by developing and using the latest technologies that thrive our resources to deliver industry-leading capabilities to our clients and customers, making it convenient for our clients to do business with InterSources Inc. Our teams also drive growth by refining technology-driven client experiences that put the users first, providing an unparalleled experience. This results in strengthening the core technologies of clients, enabling them to scale with flexibility, create seamless digital experiences and build lifelong relationships.