Lakshmanan LN
Hi, I’m Lakshmanan LN — an AI/ML engineer working on GenAI and RAG systems, MLOps, and ML architecture, writing about the systems layer underneath all of it: attention internals, LLM inference engines, and the training dynamics behind why models behave the way they do.
This blog is where I work through those topics properly — first principles first, code and diagrams over hand-waving, honest about where a technique breaks down.
Connect with me on LinkedIn, X, or YouTube to talk shop or compare notes.
Speculative Decoding: How Guessing Tokens You Then Throw Away Speeds Up LLM Inference
Here’s a claim that sounds like it can’t possibly be true: you can use a small, weak model to guess several of a large …
Read More
Agent Harnesses: Why the Same Model Scores 34 Points Apart on the Same Benchmark
Take Claude Sonnet 4.5 — same weights, same API, same day — and run it on SWE-bench Verified under two different scaffolds. Under a …
Read More
Context Engineering: Why Contexts Fail, and the Four Moves That Fix Them
In June 2025, Andrej Karpathy posted a one-line reframe that stuck: “+1 for ‘context engineering’ over ‘prompt …
Read More
vLLM Internals: PagedAttention, Continuous Batching, and Where the 2-4x Comes From
A decode step moves every weight in the model from HBM into the SMs to produce exactly one token per sequence. At batch size 1 that is a …
Read More
Scaling Laws: Kaplan, Chinchilla, and Why Nobody Trains Chinchilla-Optimal
Say you’ve got loss numbers from a 40M-parameter run and a 400M-parameter run, same data, same architecture family, same optimizer. …
Read More
Attention From Scratch: How Transformers Replaced a Bottleneck with a Lookup
Take a vanilla encoder-decoder RNN doing machine translation. It reads a source sentence token by token, updating one hidden-state vector as …
Read More