Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming

An efficient concurrent graph processing system

GoPTX: Fine-grained GPU Kernel Fusion by PTX-level Instruction Flow Weaving

Fused Triton kernels for TurboQuant KV cache compression — 2-4 bit quantization with RHT rotation. Drop-in HuggingFace & vLLM integration. Up to 4.9x KV cache compression for Llama, Qwen, Mistral, and more.

LAMB go brrr

MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.

Assigment 3 for the "Parallel & Distributed Systems" course (ECE, AUTh) - Fall 2024

Noeris — autonomous kernel fusion discovery + Triton autotuning for LLM kernels and Gemma layer deeper fusion (A100/H100 wins).

Fused Triton kernels for Transformer inference: RMSNorm+RoPE, Gated MLP, and FP8 GEMM.

Compile time kernels fusion and expression trees as Alpaka boost.odeint backend. This is my team project developed in collaboration with and under the supervision of HZDR.

High-Performance WebGPU Deep Learning Inference Engine: Zero Dependencies, Hand-Written WGSL Shaders, Kernel Fusion | 高性能 WebGPU 深度学习推理引擎

GPU fusion code and algorithm

Production-grade Triton kernel fusing residual add + RMSNorm + packed QKV projection into a single GPU launch for decoder-only transformer inference (Llama-3, Mistral, Qwen2). +2.4% tok/s, -1.5 GB VRAM on A10G.

High-performance CUDA implementation of LayerNorm for PyTorch achieving 1.46x speedup through kernel fusion. Optimized for large language models (4K-8K hidden dims) with vectorized memory access, warp-level primitives, and mixed precision support. Drop-in replacement for nn.LayerNorm with 25% memory reduction.