All work

Case study 02 / 06

Full-stack product engineering · Microfrontends · Performance

RYD Raksha — Motor Insurance Product

Owned the entire frontend of a new motor-insurance add-on product, built a reusable KYC microfrontend for other teams, and cut large reference-data payloads with Brotli, output caching and compressed browser caching.

Role
Owned frontend delivery · technical guidance to the backend team
Company
ICICI Lombard
Stack
  • Next.js
  • React
  • TypeScript
  • Zustand
  • .NET 10
  • Vite
  • Webpack
  • Brotli
  • LZ-String
  • LocalStorage
RYD Raksha: reference-data API payload compressed from about 17 MB to 6 MB with Brotli and output caching, then the browser copy from about 6 MB to 1.1 MB with LZ-String; a KYC microfrontend is embedded in other journeys.REFERENCE DATA · MAKE-MODEL + RTO① API PAYLOAD — .NET OUTPUT CACHE (ECS MEMORY)17 MB≈6 MB · BROTLI② BROWSER COPY — LOCALSTORAGE, 1-DAY TTL6 MB≈1.1 MB · LZ-STRINGREUSABLE KYC MICROFRONTEND · SCRIPT + CSSKYC widgetVITE / WEBPACK BUNDLERYD Raksha journeyMotor renewalPost-payment buying
Simplified, public-level view of the documented architecture

01

Overview

RYD Raksha is a digital motor-insurance add-on product for customers who already hold a valid third-party motor policy — including policies from other insurers. Customers choose a plan and unlock additional benefits and value-added services.

02

Context

The product needed a full customer-facing journey built quickly, with a backend team that included interns and recent graduates. Several other journeys also needed the same KYC step.

03

Problem

Deliver an interactive purchase journey end to end, avoid every team rebuilding KYC, and make large reference datasets (vehicle make-model and RTO data) cheap to download and cheap to keep in the browser.

04

Constraints

  • Tight delivery timeline.
  • Reference datasets in the tens of megabytes.
  • Browser storage quotas limit how much can be cached locally.
  • The KYC widget had to drop into applications owned by other teams.

05

My contribution

  • Owned the complete frontend flow in Next.js, TypeScript and Zustand.
  • Guided and supported backend implementation with a team of interns and recent graduates (technical guidance, not formal management).
  • Built a reusable KYC microfrontend distributed as a script + CSS bundle (Vite/Webpack) so other applications can embed a common KYC widget; it is used in the motor renewal and post-payment buying journeys.
  • Implemented Brotli compression and .NET output caching held in ECS memory for large, repeated reference datasets.
  • Implemented one-day LocalStorage caching with LZ-String compression on the client.

06

Architecture

Compress on the wire, compress at rest

Reference data is served once from .NET output cache (in ECS memory) and Brotli-compressed on the wire. On the client, the decoded data is re-compressed with LZ-String and stored in LocalStorage for one day, so repeat visits skip the network entirely.

  • Server: .NET 10 APIs with output caching for repeated datasets.
  • Transport: Brotli compression — ~17 MB to ~6 MB.
  • Client: LZ-String-compressed LocalStorage cache — ~6 MB to ~1.1 MB, 1-day TTL.
  • KYC microfrontend: Vite/Webpack build shipped as script + CSS, consumed by other journeys.
  • Live in production on the core policy system and PostgreSQL (migrated from Oracle).

07

Decisions & trade-offs

Script + CSS microfrontend instead of a shared npm package
Consuming teams embed the widget without adopting the same framework version or build pipeline.
LZ-String before LocalStorage
Browser storage quotas are small; compressing the cached copy keeps a ~6 MB dataset at ~1.1 MB so it fits.
One-day TTL
Reference data changes rarely; a day balances freshness against repeat downloads.

08

Impact

  • ~65%

    Smaller API payload

    ~17 MB → ~6 MB with Brotli

  • ~82%

    Smaller browser cache

    ~6 MB → ~1.1 MB with LZ-String

  • 1 day

    Client cache TTL

    LocalStorage

  • The KYC microfrontend is reused in the motor renewal and post-payment buying journeys.

Described at a public level — no confidential company, customer or system details.