AI

Writing on AI engineering

Notes and how-tos on building with LLMs — prompting, agents, retrieval, and evaluation. Written to help other engineers ship AI features that hold up in production.

Portrait of Sachin Gupta rendered in binary

Articles

33 articles

Writing· Search series
Fuzzy & Prefix Search
Exact match fails two ways users hit every day, typos and half-typed queries. Here is the machinery that forgives both.
Part three of the search series. The inverted index only matches literal keys, so a typo or a half-finished word returns nothing. Fuzzy search fixes typos with edit distance (Levenshtein), computed by a small dynamic-programming grid and run at scale with automata or n-grams. Prefix search fixes as-you-type with tries and FSTs. Worked examples throughout, plus an interactive edit-distance grid.
- Search
By Sachin Gupta5 Jun 20268 min read
Writing· Essay
An Agent SDK Is a Harness You Didn't Build
What a framework makes easy, and what it hides.
An Agent SDK is a packaged harness: the run loop, tool interface, memory, model abstraction, and tracing, ready to wire together. The value is what it makes easy. The risk is what it hides.
- Agent SDK
By Sachin Gupta4 Jun 20265 min read
Writing· Search series
Keyword Search: BM25
How a search engine turns a match into a score. The workhorse ranking function, and the two knobs that tune it.
Part two of the search series. Matching finds candidates; ranking decides the order. BM25 is the default text scorer, and it is really three forces: term frequency with saturation, document-length normalization, and term rarity (IDF). This walks the intuition with worked examples, the exact formula, and the k1/b knobs, then hands you an interactive score explorer.
- Search
By Sachin Gupta3 Jun 20268 min read
Writing· Search series
Text Analysis
Before a word can be indexed, it has to be prepared. The quiet pipeline that decides what your search can ever match.
Part of the search series. The inverted index does not store raw text; it stores terms, and text analysis is the pipeline that turns one into the other: tokenize, lowercase, drop stopwords, stem to a root, expand synonyms. The same analyzer must run on the query, and getting it wrong silently breaks search. With an interactive analysis pipeline.
- Search
By Sachin Gupta2 Jun 20267 min read
Writing· Search series
How Search Works
Every search engine does the same four things. Once you see them, the rest is variations.
The opening piece of a search series. Search is four steps that never change (index, parse, match, rank) and a hundred choices about how to do each. This part lays out the inverted index, boolean retrieval, why matching is not ranking, and the precision-recall tradeoff every later technique is a bet on, with worked examples throughout.
- Search
By Sachin Gupta1 Jun 20268 min read
Writing· Guide
How to Actually Evaluate an LLM System
What an eval is, the types of eval, and the step-by-step for running one that tells you something true.
Eval is how you turn a fuzzy, open-ended LLM system into something you can measure. This is a practical guide to the types of eval, from exact match to LLM-as-judge to human review, and the six-step procedure for running one you can actually trust.
- LLM Ops & Eval
By Sachin Gupta29 May 20268 min read
Writing· RAG series
Beyond the Vector
GraphRAG local and global search, the two multimodal strategies, the lineage from RETRO to CAG, each with its own analogy and visualizer.
The last piece of the RAG series: when flat vectors are the wrong shape for the knowledge. GraphRAG with local search and global search as separate ideas, multimodal RAG's two strategies (shared embedding space and translate-to-text), the family tree (RAG-Sequence, RAG-Token, REALM, Fusion-in-Decoder, RETRO, Atlas), the long-context-versus-RAG debate, and cache-augmented generation. Each with an everyday analogy and an interactive visualizer.
- RAG
By Sachin Gupta22 May 202611 min read
Writing· RAG series
RAG That Thinks
Iterative, adaptive, FLARE, Self-RAG, corrective, speculative, and agentic RAG, each with its own analogy and its own visualizer.
The control-flow axis of RAG, pattern by pattern, each with an everyday analogy and an interactive visualizer: iterative and multi-hop RAG, adaptive RAG, FLARE, Self-RAG, corrective RAG (CRAG), speculative RAG, and agentic RAG. How each adds a loop, and then adds judgment to the loop, and how to stop at the least autonomous pattern that answers your questions.
- RAG
By Sachin Gupta21 May 202612 min read
Writing· RAG series
How RAG Reads a Corpus
Fixed-size, structure-aware, sentence-window, parent-document, summary indexing, RAPTOR, and late chunking, each with its own analogy and visualizer.
The indexing axis of RAG: why chunking is the quiet make-or-break decision, and the ladder of strategies, each with an everyday analogy and interactive visualizer: fixed-size chunks, structure-aware splitting, sentence-window retrieval, parent-document (small-to-big), summary indexing, RAPTOR's hierarchical tree, and contextual and late chunking. How to decouple the unit you search from the unit you read.
- RAG
By Sachin Gupta20 May 202611 min read
Writing· RAG series
How RAG Finds Things
Sparse, dense, hybrid, learned-sparse, late interaction, re-ranking, and contextual retrieval, each with its own analogy and its own visualizer.
The retrieval axis of RAG, method by method: sparse keyword search (BM25), dense embeddings, learned-sparse (SPLADE), late interaction (ColBERT), hybrid fusion with RRF, cross-encoder re-ranking, and contextual retrieval. Each gets an everyday analogy, the real definition, and its own interactive visualizer. How each decides what counts as a match, a worked reciprocal-rank-fusion example, and why real systems combine several.
- RAG
By Sachin Gupta19 May 202612 min read

Explore by tag

AI — Sachin Gupta