Hardware + AI engineer focused on model-hardware co-design to deliver high-throughput, energy-efficient inference.

Highlights:

• 🧪 Binary Neural Networks (BNN): 1-bit models to reduce compute and memory while retaining performance.
• 🔧 High-performance model optimization: operator fusion and hardware/compiler tuning.
• ⚡️ LLM inference acceleration: quantization, kernel and operator optimization, and parallel deployment.

📙 Skills
Python C++ HLS Verilog CUDA Triton

github_local stats top-langs