view all posts
computer-vision (1)
| Median Filter over Arbitrary Datatypes |
optimization (5)
| Strassen's matmul with AVX 512 kernel | |
| Fast Matrix Multiplication with AVX 512 and Loop Tiling | |
| CUDA Histogram Kernel | |
| Dense Monocular SLAM | |
| Median Filter over Arbitrary Datatypes |
cpp (3)
| Strassen's matmul with AVX 512 kernel | |
| Fast Matrix Multiplication with AVX 512 and Loop Tiling | |
| Median Filter over Arbitrary Datatypes |
llm (5)
| Persona Vector Distillation in LLM Weights | |
| Self Doubt Interventions on Chain-of-Thought | |
| Optical Compression | |
| Confidence Aware Router | |
| LLM Routing Strategies |
efficiency (1)
| LLM Routing Strategies |
routing (2)
| Confidence Aware Router | |
| LLM Routing Strategies |
inference (1)
| LLM Routing Strategies |
ssh (2)
| SSH over Tor | |
| Setting up an SSH Server |
linux (1)
| Setting up an SSH Server |
networking (2)
| SSH over Tor | |
| Setting up an SSH Server |
server (1)
| Setting up an SSH Server |
tor (1)
| SSH over Tor |
security (1)
| SSH over Tor |
qwen (1)
| Confidence Aware Router |
vllm (1)
| Paged Attention Performance Analysis |
attention (1)
| Paged Attention Performance Analysis |
transformer inference (1)
| Paged Attention Performance Analysis |
profiling (1)
| Paged Attention Performance Analysis |
kubernetes (1)
| Kubernetes Notes |
docker (1)
| Kubernetes Notes |
devops (1)
| Kubernetes Notes |
containers (1)
| Kubernetes Notes |
bash (1)
| Updating my Bash Prompt |
prompt (1)
| Updating my Bash Prompt |
customization (1)
| Updating my Bash Prompt |
vision (1)
| Optical Compression |
compression (1)
| Optical Compression |
reasoning (1)
| Self Doubt Interventions on Chain-of-Thought |
steering (1)
| Persona Vector Distillation in LLM Weights |
distillation (1)
| Persona Vector Distillation in LLM Weights |
persona-vectors (1)
| Persona Vector Distillation in LLM Weights |
interpretability (1)
| Persona Vector Distillation in LLM Weights |
algorithm (1)
| Dense Monocular SLAM |
3d-reconstruction (1)
| Dense Monocular SLAM |
cuda (1)
| CUDA Histogram Kernel |
productivity (1)
| Fail Faster on your Ideas |
model models (1)
| Fail Faster on your Ideas |
IDE (1)
| 3 months of using Neovim |
neovim (1)
| 3 months of using Neovim |
programming (2)
| Gumbel Max Trick for Softmax Sampling | |
| 3 months of using Neovim |
sampling (1)
| Gumbel Max Trick for Softmax Sampling |
simd (2)
| Strassen's matmul with AVX 512 kernel | |
| Fast Matrix Multiplication with AVX 512 and Loop Tiling |
avx512 (2)
| Strassen's matmul with AVX 512 kernel | |
| Fast Matrix Multiplication with AVX 512 and Loop Tiling |
conviction (1)
| How to Forge Your Conviction |
mental models (1)
| How to Forge Your Conviction |
world models (1)
| How to Forge Your Conviction |
life (1)
| It takes very high agency |
philosophy (1)
| It takes very high agency |
business (1)
| 2 ways to bet on a Trillion Dollar Market |
strategy (1)
| 2 ways to bet on a Trillion Dollar Market |
anthropic (1)
| 2 ways to bet on a Trillion Dollar Market |
openai (1)
| 2 ways to bet on a Trillion Dollar Market |