Senior Software Engineer, Inference Engine (Platform Software)
New roles land in Following.
Seoul, South KoreaPosted 1 month ago
Market pay check
Matched to Computer systems and software professionals
Source: Ministry of Employment and Labor wage information system (wage.go.kr), wage distribution by occupation.
What you'd do
- Design next-generation inference engines for large and multimodal LLMs on FuriosaAI NPUs.
- Implement speculative decoding,KV-cache management,parallelism,and scheduling in production.
- Build distributed inference with prefill-decode disaggregation and hierarchical external KV storage.
What they want
- Hold a BS in Computer Science,Engineering,or related with three plus years experience.
- Demonstrate proficiency in Rust or C++ for high-performance systems and inference software.
- Show solid knowledge and interest in deep learning,LLMs,and generative AI models.
Nice to have
- Bring experience building LLM inference serving with batching,scheduling,caching,and load balancing.
- Demonstrate deep understanding of systems performance optimization for large model workloads.
- Develop kernels with C++,CUDA,or Triton for accelerator and inference performance paths.
Seoulstart's read
- Visa tier (likely)
- E-7 specialty / professional (likely)
- Korean language
- Korean not required
Related Seoulstart guides
Similar roles on Seoulstart
Software Engineer, Backend (신입) - 피드 (ML Data Platform)Daangn · SEOUL · today
Senior Software Engineer, Avatar AnimationRoblox · San Mateo, CA, United States · today