Writing

Articles

essays, notes and whitepapers on AI, vision-language models, and the systems I build

All Articles

RSS →
Aug 29, 2026
Featured
Serving LLMs: what actually decides your latency
TTFT, prefill vs decode, the KV-cache memory bill, quantization, multi-LoRA serving, and how vLLM, SGLang, llama.cpp and MLX run them across NVIDIA, AMD and Apple silicon.
27 min read·inference,serving,llm
Mar 15, 2024
Whitepapers
Beyond Imagery: AI-Enhanced Diagnostic Assistant for Cancer and Tumor Diagnosis using Radiology Imaging
Our IJETS 2024 paper — combining LLaVA-Med with Retrieval Augmented Generation to build a clinically grounded visual question answering assistant for CT and MRI imaging.