Case Study

Central Calculator Engine

A centralized financial logic engine powering financial calculators across 270+ production client applications. Simplifies maintenance by housing algorithms in a single deployable core.

Node.jsMongoDBNext.jsExpress.js

Source Access 🔒

Overview

Central Operations

00:0008:0016:0024:00

Uptime / Health

99.9%

Uptime

Applications270+

Uptime99.9%

Average< 40ms

270+Applications

< 40msAverage

99.9%Uptime

1Core

👥

270+Applications Powered

⚙️

1Core Engine (Multi-tenant)

🕒

< 40msAverage Latency

🕒

99.9%Uptime Maintained

❌

The Challenge

❌Calculators were built separately in each application as isolated components.
❌Frequent code duplication and inconsistent logic.
❌Version mismatch and calculation discrepancies.
❌High maintenance effort and deployment conflicts.

✔️

The Solution

✔️Built a centralized, highly-optimized calculation engine.
✔️Dynamic formulas and rules stored in MongoDB.
✔️Versioned engine with backward compatibility.
✔️Single deployment ensures consistency across all apps.
✔️Real-time updates without client redeployments.

Key Features

// Feature 01

Centralized formula evaluation

engine

// Feature 02

Declarative dynamic rules

in MongoDB

// Feature 03

Strict semver-based API

versioning

// Feature 04

Redis cache layer

for sub-40ms speed

// Feature 05

Multi-tenant rule deployment

safety

// Feature 06

High density microservice

design

Project Gallery

Formula Latency TrackerLIVE

Primary Stat270+ Apps Powered

Speed IndicatorAvg. Latency: < 40ms

Operations Monitor Dashboard

Mongoose Collections / Node Cache

KEY / ID	TYPE	STATUS
tax_calc_2026	Formula (Mongoose)	Deployed
mortgage_rate_eval	Processor Block	Active
sys_caching_layer	Redis Cache (TTL)	Cached

Database Collections & Cache Records

central-calculator-engine.config

1// central-formula-engine/math.ts
2import { MongoDB } from '@db/atlas'
3import { Cache } from '@redis/store'
4
5export async function calculateTax(income: number) {
6  const cacheKey = `tax:${income}`
7  const cachedVal = await Cache.get(cacheKey)
8  if (cachedVal) return Number(cachedVal)
9
10  const rules = await MongoDB.findRules('tax_calc_2026')
11  const result = evaluateRules(rules, income)
12  await Cache.set(cacheKey, result, 3600)
13  return result
14}

Source Rules Configuration Code

Project Impact

+ L00.METRIC

+ R00.STAT

+ B50.SYS

+ DATA.SEC

270+Applications Powered

1Core Engine (Multi-tenant)

< 40msAverage Latency

99.9%Uptime Maintained

Sub-40ms calculation latency maintained under peak load!

What I Learned

Versioning is Critical

Centralizing systems requires strict versioning and backward compatibility to prevent breaking client applications.

Performance is Everything

Caching, indexing, and query optimization are key to maintaining sub-40ms latency at scale.

Design for Change

A flexible rule engine and dynamic configuration layer future-proof the system for business changes.

+ CTA.01

+ END.SECTION

Open for projects

Have a similar
project in mind?

Let's build a scalable, high-performance solution from architecture to deployment.

Book a Meeting→View All Work→

let's ship it 🚀

Projects

Clients

99%