386 concepts · 19 modules · revise each in 10 min

Interview revision, not a textbook

In progress

386 concepts across JavaScript, TypeScript, React, Next.js, Laravel and AI. Each one is built to be revised in under ten minutes — definition, mental model, internals, trade-offs, cheat sheet.

Start here

Variables: var, let, const

The `var` loop bug and the React stale-closure bug are the same bug. This is where you learn to see it.

JavaScript·Lesson 1·~8 min

Begin

Lessons done

of 386

Day streak

start today

Steps read

~2 min each

Saved

bookmarks

Available

385

written so far

Modules

See the dependency map →

Module 1

JavaScript Foundations

0/28

Almost every "React bug" is a JavaScript bug in a React costume. This module is where the real leverage is.

M1 · Core MechanicsM2 · Data & FunctionsM3 · Async JavaScript

28 availableStart

Module 2

TypeScript

0/18

Types are not paperwork. They are a second program that checks the first one while you write it.

M4 · Type System FluencyM5 · Type-Level Programming

18 availableStart

Module 3

React

0/36

React is a small idea — UI = f(state) — surrounded by a large amount of consequence. We cover the consequence.

M6 · React FundamentalsM7 · Hooks MasteryM8 · Performance & PatternsM9 · State Management

36 availableStart

Module 4

Next.js

0/14

The App Router is mostly one question: where does this code run, and when? Everything else follows.

M10 · App Router

14 availableStart

Module 5

Interview Preparation

0/8

Knowing the answer and being able to say it under pressure are different skills. This module trains the second.

M11 · Production ConcernsM12 · Interview Ready

8 availableStart

Module 6

Laravel

0/30

Backend interviews are won on the request lifecycle, the container, and Eloquent — and lost on N+1 and silent failures.

M13 · Laravel FundamentalsM14 · Routing & Request HandlingM15 · Eloquent & the DatabaseM16 · Auth, Queues & AsyncM17 · Senior & Full-Stack

30 availableStart

Module 7

AI & LLM Foundations

0/23

Everything about an AI product is downstream of one fact: a language model predicts the next token. This module makes that fact — and its consequences for cost, latency and reliability — second nature.

M18 · AI & LLM Foundations

23 availableStart

Module 8

AI Application Engineering

0/16

The step between "I can call an API" and "I ship an AI product": streaming, tools, state, memory, retries and caching — the patterns every production LLM app shares.

M19 · AI Application Engineering

16 availableStart

Module 9

RAG / Knowledge Systems

0/24

Grounding the model in your data is the difference between a demo and a product. Ingestion, chunking, vectors, retrieval, reranking — and the evaluation that tells you it works.

M20 · RAG / Knowledge Systems

23 availableStart

Module 10

AI Agents

0/19

An agent is a loop with tools. The architecture is the loop — perception, decision, action, observation — and everything else is the discipline that keeps it safe, bounded and observable.

M21 · AI Agents

19 availableStart

Module 11

AI Automation

0/16

Where AI earns its keep: workflows that touch the CRM, the inbox, the database and the API layer — with queues, approval gates and a recovery story.

M22 · AI Automation

16 availableStart

Module 12

Backend & Distributed Systems for AI

0/28

The model is the easiest part of the backend. Auth, rate limits, queues, caching, streaming transport and fault tolerance are what make an AI product a platform.

M23 · Backend & Distributed Systems

28 availableStart

Module 13

Cloud & AWS for AI

0/27

IAM, Lambda, Bedrock, SQS, Step Functions — the AWS vocabulary an AI architect is expected to design in, and the cost controls that keep the bill sane.

M24 · Cloud & AWS for AI

27 availableStart

Module 14

Docker / DevOps / Infrastructure

0/20

The pipeline that ships an AI service: containers, CI/CD, IaC, canaries and rollbacks — enough Kubernetes to speak the language, not to run the cluster.

M25 · Docker / DevOps / Infrastructure

20 availableStart

Module 15

AI Security

0/20

Prompt injection, excessive agency, poisoned knowledge bases — the AI attack surface is new, and the OWASP LLM Top 10 is the map. This module is the defense in depth.

M26 · AI Security

20 availableStart

Module 16

AI Observability & Evaluation

0/19

You cannot improve what you cannot measure. Tracing, token metering, cost attribution, groundedness checks and the eval suite that runs in CI like tests do.

M27 · AI Observability & Evaluation

19 availableStart

Module 17

AI System Design

0/12

The round that decides the architect: run any AI prompt — chat, RAG platform, support, sales, recruiting, coding assistant — through one repeatable protocol.

M28 · AI System Design

12 availableStart

Module 18

Enterprise AI Solutions Architecture

0/22

The architect half of the role: requirements, stakeholders, ADRs, build vs buy, cost models, governance and compliance — the deliverables that are not code.

M29 · Enterprise AI Solutions Architecture

22 availableStart

Module 19

Capstone Projects

0/6

Six production-grade projects — a RAG SaaS, a guarded agent, an automation platform, a multi-tenant AI SaaS, an enterprise assistant and one complete case study. Serious projects, not toy chatbots.

M30 · Capstone Projects

6 availableStart

Highest interview frequency

If you only have an hour

100Top JavaScript Interview QuestionsInterview 100Top TypeScript Interview QuestionsInterview 100Top React Interview QuestionsInterview 100Top Next.js Interview QuestionsInterview 96useEffectReact 95ClosuresJavaScript

Achievements

0/10

First Step

Momentum

Consistency

Habit

JavaScript

TypeScript

React

Next.js

Laravel

Interview Ready