Home/Case Studies/Karat
CASE STUDYFINTECH · AA FINTECH DATA PIPELINE

Karat

A goals-driven investment platform that aggregates live bank, mutual fund, and equity data, profiles investor behaviour, and turns net worth into clear, actionable insights.

121Weeks operated
6k+Hours of work
8Engineers

The brief

Build a goals-driven investment platform that helps retail investors understand the relationship between time and money, optimise financial decisions through the SLICE model, and gain insight into their portfolio through behavioural profiling, net worth aggregation, and actionable financial narratives — starting with a Supabase-backed MVP and evolving into a production AWS-native platform.

What we built

A full-stack B2C fintech platform branded as "time.money" / Karat Capital — a goals-driven investment platform implementing the SLICE model (Sustainable Investment Strategy, Liquidity, Implied Volatility Lower, Consistency, Efficiency). The platform comprises:

  • an investor onboarding flow with passwordless Cognito auth and PAN verification
  • an Account Aggregator consent pipeline fetching live financial data (banks, mutual funds, equities, ETFs) via Digitap and CAMS
  • behavioural and money quotient assessments via Syntoniq
  • a Net Worth + Investor Narrative Dashboard with hierarchical financial calculations, credit score, 7-70 Winning Formula, and Goals Builder module
  • subscription billing via Razorpay
  • an admin/internal dashboard in SvelteKit; and
  • two Astro-based marketing websites (karatcapital.com, time.money).
⊛ Live in production

Production platform live at app.time.money and karatcapital.com with: full AWS Cognito passwordless auth, Account Aggregator onboarding (CAMS + Digitap), Behaviour Quotient + Money Quotient assessments, Investor Narrative Dashboard (net worth, market portfolio, loans, assets, credit score, 7-70 Winning Formula), Goals Builder with retirement projections, Razorpay subscription billing (Discovery/Advisory plans), DPDPA consent manager, and PostHog analytics.

How it was built, phase by phase.

8 workstreams across 121 weeks of operated delivery

  1. (001)BUILD

    Supabase + Deno/Oak Backend Foundation (Phase 1)

    Initial backend architecture using Supabase for auth and database, Prisma ORM for migrations, and Deno/Oak as the server runtime. Included phone OTP, email verification, SSO redirect, and onboarding schema design.

    Phase-1 backend scaffolding with auth, phone OTP, email verification, SSO, and schema migrations established.

    SupabasePrisma ORMDenoOakNode.jsNodemailer
  2. (002)BUILD

    AWS-Native Backend Architecture (Spring Boot + DynamoDB + Cognito)

    Complete migration to AWS-native stack: Spring Boot (Java 21, Amazon Corretto) as the microservice layer, DynamoDB with AWS Encryption SDK for at-rest data security, AWS Cognito for passwordless auth.

    Production-grade AWS microservice architecture with encrypted financial data storage, event-driven processing, and feature-flagged rollouts.

    Spring BootJava 21Amazon CorrettoAWS DynamoDBAWS Encryption SDKAWS Cognito
  3. (003)INTEGRATE

    Account Aggregator Integration (Digitap + CAMS)

    Full integration with India's Account Aggregator ecosystem via Digitap and later CAMS (CAMSfinserv).

    End-to-end AA consent-to-data pipeline live, supporting banks, mutual funds, equities, deposits, and ETFs with encrypted storage.

    Digitap AA APICAMS AAAnumatiAWS SQSAWS EventBridgeAWS DynamoDB
  4. (004)BUILD

    Financial Data Modelling & Credit Analytics (Experian/CIBIL)

    Schema design for credit analytics, CIBIL/Experian consent flows and data processing, bank data parsing, and a Net Worth calculation engine using Composite design pattern across assets/liabilities/market portfolio.

    Hierarchical net worth calculation API covering all asset classes (banks, MF, equities, ETFs, insurance, loans) with credit score integration.

    AWS DynamoDBSpring BootAWS API GatewayExperian APIDigitapComposite Pattern
  5. (005)BUILD

    Next.js 15 Frontend with Design System (DaisyUI + TailwindCSS)

    Full React/Next.js 15 frontend architecture with DaisyUI component library, TailwindCSS v4, custom design system (typography, colors, tokens), dark/light theming, AG Grid for financial tables.

    Production-grade financial dashboard UI with design system, dark mode, responsive layouts, and rich data visualizations.

    Next.js 15React 19DaisyUITailwindCSS v4AG GridECharts
  6. (006)INTEGRATE

    Behavioural & Money Quotient Assessment Engine (Syntoniq)

    Integration of Syntoniq's behavioural questionnaire API for the 'OhBehave' (Behaviour Quotient) and Money Quotient assessments.

    Full BQ + MQ assessment engine live with server-side question fetching, dynamic question types, and score-driven narrative generation.

    Syntoniq APISanity CMSAWS DynamoDBSpring BootNext.jsReact Hook Form
  7. (007)BUILD

    Investor Narrative Dashboard & Financial Insights

    The core investor-facing dashboard showing net worth, market portfolio, assets, loans, credit score (Experian), 7-70 Winning Formula, Behaviour Quotient, and portfolio suitability.

    Full investor dashboard live with net worth, market portfolio, credit score, insights, goals builder, and SLICE-model-aligned narrative.

    Next.jsSpring BootEChartsGauge.jsAG GridSanity CMS
  8. (008)BUILD

    Auth & Identity (AWS Cognito Passwordless + Session Security)

    Production auth system using AWS Cognito with passwordless phone/email OTP, AWS Amplify, protected route middleware, idle session timeout, JWT token refresh, cookie encryption, AWS Verified Permissions POC.

    Hardened auth system with passwordless login, idle timeout, token refresh, WAF bot protection, and DPDPA-compliant cookie handling.

    AWS CognitoAWS AmplifyAWS Verified PermissionsAWS WAFCloudFrontNext.js Middleware

Stop renting hours. Start running functions.

Pick the function you want off your plate. We map the brain and name the outcome before you commit.