AI LEADER • VISUAL DESIGN ENTHUSIAST
A software engineer with a balanced left and right brain, specializing in pipeline development, DevOps, graphics tools, and site reliability.
Inside vLLM's day-0 support for NVIDIA Vera Rubin NVL72: 19.16 TB/s HBM4 bandwidth, 50 PFLOPS NVFP4 Tensor Cores, CUDA 13.4 locality domains, and 7.8x throughput scaling over GB200.
Read Post →Dissecting OpenAI DevDay 2026: persistent Dots agents on cloud computers, GPT-6.1 Sol 80% price collapse, typed Decisions API, Salesforce Koa CRM reasoning, and SGLang NVFP4 KV cache.
Read Post →Inside vLLM Semantic Router's Decision 3.0 (d3): 5 open models (0.85B to 27B) unlocking multimodal System One decision scoring across text, high-res images, and video streams in a single non-autoregressive forward pass.
Read Post →
