Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: NLP
Why RAG Exists: Parametric Memory, Dense Retrieval and the 2020 Paper
Why RAG exists: the limits of parametric memory, how BM25 and embeddings led to dense retrieval, and what RAG-Sequence and RAG-Token trade off in the paper.
7 min readAuditing Order Confirmations an Agent Never Placed: Checking Values Instead of Wording
How to build a claim-consistency evaluator for a support agent that places real orders: partition outcomes by owner, verify identifiers against tool results.
16 min readData Cleaning for Production RAG: Encoding, OCR, Dedup and PII
How to turn parsed enterprise documents into retrieval-ready text: encoding repair, boilerplate and duplicate removal, risky OCR flags, language handling and PII masking.
21 min readHow RAG Cuts LLM Hallucinations: Retrieval, Embeddings and Enterprise Use
Why LLMs answer wrong with confidence and how RAG grounds them in current data: the two-part architecture, embeddings, fine-tuning trade-offs and limits.
9 min readFrom Rules to Transformers: How NLP Learned to Handle Context
A short history of natural language processing: rule-based systems, N-grams, Markov models, word vectors, RNNs, attention, Transformers and the large language models built on them.
7 min readWhy Transformers Use Sine and Cosine for Positional Encoding
A step-by-step explanation of sinusoidal positional encoding, from why attention ignores word order to clocks at different speeds, a worked 12-dimension example and rotation.
8 min readBuild a Transformer From Scratch in PyTorch: Attention, Encoder and Decoder
Step-by-step PyTorch code for the core Transformer pieces: positional encoding, scaled dot-product and multi-head attention, feed-forward layers, encoder stack and decoder basics.
8 min readWhy Indian Languages Cost More Tokens in LLMs and How Tokenizers Get Fixed
The same sentence takes 5 tokens in English and 34 in Tamil. See how BPE training data causes it, what rebalancing and vocabulary size change, and how vocabulary expansion works.
8 min readHarness Engineering: The Foundation of Reliable AI Agents
Harness engineering as the foundation of reliable AI agents: what an LLM is, what makes it an agent, and what the harness adds around the agent loop itself.
11 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.