1d ago · 14 min read · Introduction AI agents are no longer experimental curiosities, they’re core components of production systems, orchestrating workflows, making autonomous decisions, and interacting with users in real t
Join discussion2d ago · 14 min read · System design has always been a critical skill for developers. When we talk about traditional system design, we usually discuss how to design scalable applications such as food delivery apps, payment
Join discussion
6d ago · 7 min read · TL;DR After del tensor; torch.cuda.empty_cache(), PyTorch’s caching allocator still holds 53.7 MB that it won’t release. We traced the CUDA Runtime and Driver APIs with eBPF uprobes to see exactly wh
Join discussion
May 26 · 6 min read · Three eBPF patterns hyperscalers run in production today, mapped to the equivalent patterns on the GPU plane that nobody runs in production yet. TL;DR GitHub recently disclosed using eBPF in productio
Join discussion
May 26 · 7 min read · Repo: ashutosh-rath02/llm-gateway Why Build Another Layer Around LLMs? Most teams do not begin with a gateway. They begin with a single feature, a provider SDK, one prompt, and a direct API call insi
Join discussion
May 21 · 6 min read · TL;DR PyTorch’s DataLoader can be 50-124x slower than direct tensor indexing for in-memory GPU workloads. We reproduced a real PyTorch issue on an RTX 4090 and traced every CUDA API call and Linux ke
Join discussion
