Skip to content
Finance Archive

About the archive

Built to make finance click.

Finance Archive is an interactive finance library for students and curious professionals. Every article follows the same method — learn the idea, calculate with it, then apply it — and the platform adds a formula library, practice, sample exams, personal collections, collaboration, and a hands-on Finance Lab.

The method

  • LearnA clear explanation of the idea, the formula, and why it matters in practice.
  • CalculateAn interactive tool. Change the assumptions and follow the result for yourself.
  • ApplyA decision case and self-check questions that turn the idea into judgement.

Coverage

A library that is already deep.

Everything below is live content, wired to the same records the interface renders.

25

Structured articles

76

Formulas

90

Practice questions

31

Concept records

5

Curricula

5

Sample exams

4

Lab cases

13

Subject domains

How it is built

Principles before features.

The archive is small on the surface and strict underneath, so new topics and tools drop in without rework.

One topic, one recordEvery article, concept, formula, and question is a typed data record. The interface is a view over that record — never the source of truth.
The interface never calculatesAll arithmetic lives in a React-free finance engine with its own tests. Charts and pages display results; they never invent them.
AI explains, it never computesAny generated guidance is bounded to plain-language explanation. Numbers always come from the deterministic engine.
Demo-first, database-optionalThe archive runs with zero configuration from typed seed data, then switches to MongoDB the moment a connection string is present.

Documentation

Every system has a written spec.

The repository ships with the documentation below. Each file explains one part of the platform, from the design tokens to the finance engine and the database.

OverviewREADME.mdProject overview, quick start, routes, and structure.
Architecturedocs/ARCHITECTURE.mdLayering, data flow, and the content read boundary.
Design systemdocs/DESIGN-SYSTEM.mdColor, type, spacing, motion, and responsive rules.
Content modeldocs/CONTENT-MODEL.mdThe note type and every supporting collection.
Authoring guidedocs/AUTHORING.mdHow to add an article in the studio or in code.
Knowledge modeldocs/KNOWLEDGE-MODEL.mdConcepts, typed relations, and ordered curricula.
Learning enginedocs/LEARNING-ENGINE.mdScoring, mastery bands, and recommendations.
Finance enginedocs/FINANCE-ENGINE.mdStatements, metrics, valuation, and risk.
The Finance Labdocs/LAB.mdThe experiment loop and every Lab module.
The Ledgerdocs/LEDGER.mdThe editorial desk: model, routes, and authoring.
Company simulatordocs/SIMULATION.mdThe deterministic month-by-month simulator.
Visualizationdocs/VISUALIZATION.mdFormatting, charts, statements, and the graph.
The AI layerdocs/AI.mdWhere assistance is allowed and what it must never do.
API referencedocs/API.mdEndpoints, response envelope, and indexes.
MongoDB & seedingdocs/MONGODB-SEEDING.mdThe data layer, seed plan, and recommended indexes.
Assessmentdocs/ASSESSMENT.mdImplementation assessment and the phased plan.
ValidationVALIDATION.mdWhat every automated check covers, and why.
Learning experienceLEARNING-EXPERIENCE.mdThe Learn / Calculate / Apply experience spec.
Case studydocs/CASE-STUDY.mdProblem, method, architecture, and results.

Case study

From scattered notes to a system.

A short version of the project write-up, with the full engineering story on its own page.

The problem

Finance notes are piles, not paths

Students collect scattered notes and formulas with no sense of order, no way to check understanding, and no place to use an idea once it is read.

The approach

One method, enforced by the data model

Every article is a Learn / Calculate / Apply document. A typed knowledge graph links concepts, relations, and curricula so the archive stays navigable as it grows.

The outcome

A complete, tested learning platform

A responsive archive, a formula library, practice and exams, personal collections, collaboration, and a deterministic Finance Lab — all driven by the same records.

View the working result in the Finance Lab, or start from the archive.

ZH

Creator & maintainer

Zisanur Haque

Who builds this

Designs and builds the archive — the content model, the finance engine, and the interface. The platform is designed as a durable teaching tool: consistent structure, verified calculations, and documentation for every moving part, so the archive can keep growing without losing its shape.

Questions, corrections, or a topic you would like covered? Reach out through the links here, or read how to author an article yourself.