GPU/AI Application Platform Engineer Intern (Server Platform) – 2026 Fall (BS/MS)
About this internship
TikTok is hiring a GPU/AI Application Platform Engineer Intern (Server Platform) for Fall 2026 in San Jose, CA, working on-site. This is a 12-week, full-time internship, paid $43–$57 per hour. You’ll likely apply your understanding of computer system architecture—especially GPU/AI SoC or platform architecture, interconnect fabric, and memory subsystems—to help improve platform architecture and application performance.
Qualifications
Required
- Be able to commit to a 12-week full-time work period during Fall 2026
- Thesis in GPU/AI platform architecture and/or application performance optimization design or software hardware co-design.
- Deep understanding of computer system architecture, especially on GPU/AI SoC or Platform Architecture, Interconnect Fabric, and Memory sub-system.
Nice to have
- Coursework or research experience related to GPU/AI platform architecture, interconnects, or memory hierarchy.
- Experience analyzing application performance and proposing practical optimization ideas based on architectural bottlenecks.
- Background working across hardware and software considerations, such as software-hardware co-design in a project or thesis context.
- Ability to clearly document technical findings and communicate tradeoffs in architecture or performance decisions.
- Comfort collaborating with engineers and iterating on designs based on feedback and testing results.
How to stand out (TUN tip)
Before applying, prepare a short, specific summary of your thesis or projects that connects directly to GPU/AI platform architecture, interconnect fabric, memory subsystems, or performance optimization. If you have any write-ups, diagrams, or benchmarks from your work, consider linking them in your application materials. Confirm you can commit to a 12-week, full-time on-site schedule in San Jose during Fall 2026.