Pinned Loading

1. forgecc-tensor-compiler forgecc-tensor-compiler Public

   ForgeCC is a C++20 mini tensor compiler that transforms a custom tensor language into native CPU executables through parsing, semantic analysis, graph and loop IRs, optimization, memory planning, C…

   C++

2. inferserve-low-latency-inference inferserve-low-latency-inference Public

   A C++20 inference server built around tail latency. Incoming requests are queued, batched dynamically, and run on a GPU through ONNX Runtime or TensorRT. Ships with an open-loop load generator that…

3. kernellab-cuda-optimization kernellab-cuda-optimization Public

   Hand-written CUDA kernels for the operations that dominate deep learning: GEMM, softmax, layernorm, and attention. Each kernel is built up in steps, from a naive version to a tuned one, and compare…

4. my-certificates my-certificates Public

   A well-organized collection of my verified certificates and badges across Web Development, AI/ML, Cloud, and other technical domains — showcasing my growth, skills, and dedication to continuous lea…

   5