Skip to content
Skip to the lesson
← RoadmapDay 74 of 90AI Engineering2h 30m

Chunking, indexing, and retrieval

By the end of today you can turn a set of documents into something searchable by meaning, and you can explain why chunking decisions determine retrieval quality more than the choice of vector database ever will.

YesterdayOn Day 73 you compared two pieces of text. Today you do it across thousands, which means deciding what a piece is.

TomorrowTomorrow you feed what you retrieved into a model, and meet every way that pipeline fails.

01

Why this matters

Almost every failing retrieval system fails at chunking, not at the model. Splitting a document badly makes the right answer unfindable no matter how good everything downstream is.

  • Chunking
  • Vector stores
  • Retrieval and cutoffs
  • Hybrid search
Free tool for todayChunking playgroundSee exactly where your document splits and what each chunk costs.
02

Learn it

75 min

Copy this into Claude or ChatGPT. It quizzes you before it explains anything, which is deliberate. The resources under it are how you check what it told you.

Today's Master Prompt

Free · sign in

A prompt written for this day alone: your level, the exact scope, what to leave out, and an instruction to quiz you before it explains anything. Paste it into Claude or ChatGPT and it teaches you today's material.

Sign in to continueNo card, now or later.

Check it against something that is not a model

An assistant can be fluent and wrong, and on a topic you met today you will not catch it. These cover the same ground and were made by people who do this for a living, so they are what you hold the explanation up against. They are other people's work and we only link to them, so judge them for yourself.

3 hand-picked resources

Free · sign in

Videos, official docs and articles covering the same ground, each opened and annotated by hand. They are what you check the assistant against on a day you cannot yet catch it being wrong.

Sign in to continueNo card, now or later.
03

Build it

55 min

Take a real document set of at least twenty pages. Build the full pipeline: parse, chunk, embed, index. Write ten questions you know the answers to. Measure how many are answerable from the top three retrieved chunks. Then change chunk size dramatically in both directions and measure again. Record all three numbers.

04

Recall it

20 min

Answer out loud, reveal, then mark honestly whether you had it. That score is the only thing on this page you do not get to choose.

5 recall questions

Free · sign in

Questions you answer from memory, then grade yourself against the real answer. The score is carried into the mastery rating below it, so an honest miss cannot quietly become a tick.

Sign in to continueNo card, now or later.
05

Rate it

Completion and mastery are tracked separately. Be honest, because an inflated rating only means the concept resurfaces sooner.

Mastery tracking

Free · sign in

Rate yourself against five named criteria per concept. Completion and mastery are tracked separately, and anything you rate shakily comes back automatically on a spaced schedule.

Sign in to continueNo card, now or later.
06

Recap

  • 01Bad chunking makes the right answer unfindable, whatever comes after
  • 02Small chunks are precise and contextless; large ones are contextual and vague
  • 03Metadata filtering often beats a better embedding model
  • 04Most projects do not need a dedicated vector database

Your progress

Free · sign in

Mark days complete, pick up where you left off across devices, and watch completion and mastery diverge. Free, and the account exists only so ninety days of work cannot vanish with a cleared browser.

Sign in to continueNo card, now or later.