Take part in a hands-on internship focused on enhancing the performance of core software components for AI hardware. Based in Belgrade, this 3-6 month onsite role offers the chance to work directly with experienced engineers on optimizing kernels that power AI acceleration, particularly around convolution operations and tensor computations.
What You'll Do
- Design, implement, and refine software components tied to AI hardware performance
- Optimize low-level code for efficiency, with a focus on convolution routines and tensor operations
- Use performance profiling tools to detect and eliminate bottlenecks
- Collaborate with machine learning teams to integrate optimized kernels into broader frameworks
- Write and maintain unit tests to ensure reliability and stability of kernel-level code
- Present your technical approach, progress, and results at the end of your internship
Who You Are
You're currently in your final year of a bachelor’s or master’s program in Computer Science, Engineering, Mathematics, or a related field. You have working knowledge of C++ or Python and a solid grasp of algorithms and data structures. You're drawn to low-level programming and eager to explore kernel development, performance tuning, and optimization techniques.
Familiarity with machine learning concepts and frameworks is a plus. You must be able to commit to at least three months, with availability for up to six. Due to export control regulations, candidates must be eligible to access U.S. export-controlled technology, which may depend on citizenship, residency status, or licensing approval.
What We Offer
- Competitive compensation and mentorship from industry professionals
- Real-world experience in RISC-V architecture and AI hardware-software co-design
- Exposure to high-impact challenges in AI acceleration
- Potential for extension based on performance and project needs
Work Environment
This is an onsite position in Belgrade, Serbia. We value curiosity, collaboration, and a drive to solve difficult technical problems. Our culture centers on building the most advanced AI platform through open, efficient computing.
Tenstorrent is an equal opportunity employer. We welcome applicants from all backgrounds and experiences.
