Hi, I'm Ashish Goyal

I'm a Machine Learning Engineer specializing in Large Language Models and LLM inference optimization. Currently at Verloop.io, I'm building autonomous customer support systems powered by LLMs to serve enterprise clients at scale.

I believe in understanding AI systems at a fundamental level—from transformer architectures to inference optimization. Through my technical blog, I break down complex topics like Rotary Positional Embeddings (RoPE), attention mechanisms (MHA, GQA, MLA), and modern LLM architectures, providing complete PyTorch implementations from scratch.

My work bridges cutting-edge research and production systems. I'm currently diving deep into LLM inference optimization — experimenting with serving runtimes like vLLM, TensorRT-LLM, and SGLang, and exploring techniques like quantization, speculative decoding, and KV cache optimization to serve models at scale efficiently. I focus on building with clarity, sharing knowledge openly, and making advanced ML techniques accessible to practitioners.