CUDA Developer - C/C++ Programming (5-15 yrs)

Role : CUDA Developer

Location : Anywhere in the World - Permanent Remote

Experience : 5 - 8 Years

Role : Contract/Duration: 6 -12 Months

Opportunity : Full Time, 8 hours

Overlap Required : 4 hours PST timing

Work timings which is : Primarily, overlap is required from 8 AM to 12 PM PST. The remaining 4 hours can be accommodated either before 8 AM PST or after 12 PM PST

About the Role :

A U.S.-based company that is creating an all-encompassing AI content creation and management system for modern GTM teams, is looking for a Full stack engineer with experience in building products in the telephony/VoIP space. The company is helping businesses to create trustworthy, clear-cut, and effective AI-based solutions that foster marketing innovation. The company has managed to secure more than $20mn during its Series A round of funding.

Job Description :

The ideal candidate should have a strong background in parallel computing and performance optimization using CUDA.

Must Have :

Total Years Of exp- 5+ years experience mandatory

Mandatory Skills: C++ - 5 years, LLM - 2 years , 2 years exp in CUDA development.

Responsibilities :

- Identify and resolve performance issues and bugs related to CUDA usage.

- Analyze and improve algorithm performance through efficient parallelization.

- Stay updated with new technologies and best practices in CUDA and parallel computing.

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 development.

- Ability to solve complex problems and work both independently and in a team.

- Good communication skills and ability to work in a collaborative environment.

Knowledge and Skills :

- Proficiency in C/C++.

- Experience with code optimization for performance and efficient resource usage.

- Experience with CUDA version 12.0 or above (12.3 preferred).

- Knowledge of frameworks and libraries like cuBLAS and cuDNN.

- CUDA Performance Optimization.