DEV Community: jemst7480-del

How I Built an Actuarial Mortgage Platform with 163 Pages and Zero Technical Debt

jemst7480-del — Sun, 25 Oct 2026 12:51:15 +0000

When shopping for a home, most consumers rely on standard bank mortgage calculators. Unfortunately, the vast majority of online mortgage widgets are mathematically oversimplified: they calculate a clean Principal & Interest payment, while completely hiding private mortgage insurance (PMI) cancellation curves, tiered transfer taxes, and front-loaded interest mechanics.

To solve this, I spent the last few months architecting SmartMortgageCalc, an open actuarial intelligence suite designed to deliver institutional-grade calculations with zero data tracking.

Here is an architectural breakdown of how the platform is engineered to handle 163 prerendered pages, cross-border banking regulations, and pass 29 rigorous actuarial test suites.

1. The Core Engineering Challenge: Beyond $M = P \frac{r(1+r)^n}{(1+r)^n - 1}$

The standard amortization formula is elementary:

$$M = P \frac{r(1+r)^n}{(1+r)^n - 1}$$

Where:

$M$ = Monthly principal and interest
$P$ = Loan principal
$r$ = Monthly interest rate
$n$ = Total number of payments

However, in the real banking world, no mortgage is this simple:

Cross-Border Compounding Rules:
- In the United States, interest compounds monthly.
- In Canada, under the Canada Interest Act, fixed mortgages must compound semi-annually, not monthly. A nominal 5.0% Canadian loan produces a different effective amortization rate than a 5.0% US loan.
- In Australia, offset accounts recalculate interest on daily balances, and biweekly payments follow an accelerated 26-pay schedule that saves thousands in compound interest.
The Front-End DTI Benchmark:
- Under standard 28/36 underwriting guidelines, a buyer purchasing a $400,000 property needs an exact salary calculation factoring in property taxes and homeowner insurance. We documented this full breakdown in our Salary Needed for a $400k House Guide.

2. Testing for Precision: 29 Actuarial Test Vectors

To guarantee zero mathematical drift, every underwriting equation is backed by automated Node.js test suites.

Here is a snippet from our automated test suite evaluating early mortgage payoff and principal curtailment:


typescript
// Benchmark: $300,000 loan at 6.00% fixed for 30 years
// Baseline Monthly P&I = $1,798.65
// Testing +$200/month extra principal curtailment
const loanAmount = 300000;
const annualRate = 0.06;
const extraPayment = 200;

const result = calculateAcceleratedPayoff(loanAmount, annualRate, 360, extraPayment);

assert.strictEqual(result.yearsSaved, 6.8);
assert.strictEqual(Math.round(result.interestSaved), 91230);

Why I Built a Mortgage Calculator That Doesn't Track You (100% Client-Side)

jemst7480-del — Wed, 21 Oct 2026 16:13:27 +0000

As developers, we build tools to solve problems. But when I was looking for a mortgage calculator for my own personal use, I noticed a disturbing trend: almost every calculator out there requires you to sign up, give your email, or worse, they track your financial data and sell it to third parties.

That didn't sit right with me. Financial data is deeply personal.

So, I decided to build my own: Smart Mortgage Calc.

Here is why I built it, the tech stack I used, and the challenges of building financial tools in the browser.

The Core Problem: Privacy

When you input your income, your down payment, and your desired loan amount into a standard calculator, where does that data go? Usually, it's sent to a server, stored in a database, and used for targeted advertising.

I wanted to build a tool where the user has absolute trust. The data never leaves your browser.

The Tech Stack

To achieve 100% client-side processing, I kept the stack simple, lightweight, and fast:

Vanilla JavaScript: No heavy frameworks like React or Angular. Just pure, fast JS to handle the math.
HTML5 & CSS3: For a clean, responsive UI that works perfectly on mobile (since most users access financial tools on their phones).
No Backend (for calculations): The math (amortization, extra payments, PMI calculations) is entirely handled by the client's browser.

Key Features Built from Scratch

Standard Amortization Schedule: A full breakdown of principal vs. interest over 30 years.
Extra Payment Simulation: I wanted users to see exactly how much they save by paying an extra $100 or $200 a month. The math for this is complex, but the visualization is powerful.
Discount Points Break-Even Analysis: A feature many people don't understand, so I built a tool to show exactly when buying points makes financial sense.

The SEO & Performance Challenge

Building a financial tool is one thing; getting it discovered is another. Because the app is client-side, I had to rely heavily on Semantic HTML and Schema Markup to help Google understand what the tool does. I also wrote in-depth blog posts to answer specific user questions (like "How much income do I need for a 400k house?").
https://smartmortgagecalc.space/extra-payment

Try It Out

If you are buying a home, refinancing, or just love personal finance, you can check it out here:

👉 smartmortgagecalc.space (No sign-ups, no cookies, just pure math).

I would love to hear your feedback on the UI, the performance, or how I can improve the calculations!

javascript #webdev #privacy #fintech #showdev

How I Built a Free Mortgage Calculator with 100% Client-Side Privacy

jemst7480-del — Tue, 29 Sep 2026 11:16:34 +0000

I built SmartMortgageCalc.space because I was tired of mortgage calculators that ask for your email, sell your data, or hide the real numbers behind lead forms.

The goal was simple: build a calculator that runs 100% in the browser, with no server-side data storage, and no sign-ups.

The Tech Stack:
• Google AI Studio (Gemini) for initial development
• Google Cloud Run for hosting
• Cloudflare for DNS and security
• Vanilla JavaScript for calculations
• Chart.js for visualizations

The Privacy Challenge:
Most mortgage calculators send your financial data to a server. I wanted to avoid that. The solution: all calculations run client-side using standard amortization formulas. No data leaves the browser.

The SEO Challenge:
I needed the site to be discoverable. I implemented:
• Schema.org structured data (WebApplication, FAQPage, HowTo)
• Dynamic sitemap generation
• Core Web Vitals optimization (LCP < 1.5s)
• Internal linking architecture

The Result:
A free, fast, privacy-first mortgage calculator that works for US, Canada, and Australia.

If you're building a tool that handles sensitive data, consider client-side calculations. It's better for users, and it's better for trust.

Check it out: https://smartmortgagecalc.space