triton , Noc, Verilog, DataFlow, Tiling, LLM workloads, AI inference chips, memory hierarchy, DMA scheduling, hardware-software co-design, TVM, compiler-guided optimization, XLA, VHDL, memory planning, LLVM, Chisel, kernel optimization tools, MLIR, systemverilog, microarchitectures, AI accelerator architectures, operator mapping, compiler frameworks, compute arrays