Serbia- AI Workloads Engineer
Core
Build workflow modeling infrastructure and adapt open-source AI systems to drive performance improvements on NextSilicon's next-generation HPC and AI hardware platforms.
Role type
Senior IC AI Workloads Engineer
Builds
Workflow modeling infrastructure and optimized AI inference systems
Domain
High-performance computing (HPC) and AI inference
Deliverable
production ML models
Required skills
Python, PyTorch, C++, LLM inference workflows, distributed inference, performance analysis, workload modeling
Preferred skills
Dynamic batching systems, MLPerf Inference benchmarks, custom kernel programming (CUDA, Triton), parallelism strategies (TP/PP/EP)
Technologies
vLLM, TensorRT-LLM, NCCL, CUDA, Triton
Responsibilities
Model and analyze end-to-end AI workflows from simulation through production; Run and adapt open-source AI workloads to collect metrics; Profile full-stack behavior to identify bottlenecks; Prototype custom kernels or runtime components for optimization.
