Skip to content
All case studies

Invincible Ocean Pvt Ltd · Jun 2023 – Dec 2024

A secure, fast API platform for fintech products

Built and hardened the API layer behind fintech products: 350+ APIs, tiered access control, protected storage and a much faster data layer.

Full-stackDeployment & DevOps

350+

Public & in-house APIs architected

40%

Faster MongoDB response times

40%

Less downtime from CronJob orchestration

25%

Better server performance

01Context

Invincible Ocean builds fintech products for B2B and B2C users. Over 19 months I grew from Associate Software Developer to Software Developer, working on the backend that its client-facing products depend on.

02Challenge

The products needed a large and reliable API surface for clients and internal tools, access control that separated admins from users, file storage that was never publicly exposed, and response times that held up as usage grew.

03Approach

  1. 01

    Build the API surface

    Architected and developed 350+ public and in-house APIs, and tested them across edge cases with Postman before they reached clients.

  2. 02

    Lock down access

    Engineered role-based access control with three tiers (Super Admin, Admin, User) on top of JWT authentication, plus a customisable IP whitelist so clients could restrict where calls come from.

  3. 03

    Keep files private

    Moved storage to private S3 buckets. Files are served through short-lived pre-signed URLs, so nothing is publicly addressable.

  4. 04

    Make it fast

    Added Redis caching in front of hot API responses and optimised MongoDB queries and indexes using Compass and Studio3T.

  5. 05

    Automate operations

    Set up Jenkins CI/CD for repeatable deployments, orchestrated scheduled CronJobs, and managed the EC2 and S3 infrastructure behind it all.

04Architecture

  1. 1Consumers

    • Client apps
    • In-house tools
  2. 2Gatekeeping

    • IP whitelist
    • JWT auth
    • RBAC · 3 tiers
  3. 3API layer

    • 350+ REST APIs
    • Scheduled CronJobs
  4. 4Data

    • Redis cache
    • MongoDB
    • S3 · pre-signed URLs

Simplified view of how the pieces connect.

05Stack

REST APIsJWTRBACRedisMongoDBAWS S3AWS EC2JenkinsCronJobsPostmanStudio3T

06Outcomes

  • 350+ · Public & in-house APIs architected
  • 40% · Faster MongoDB response times
  • 40% · Less downtime from CronJob orchestration
  • 25% · Better server performance
  • Three-tier RBAC with JWT across the platform
  • Private S3 storage served only through pre-signed URLs
  • Client-configurable IP whitelisting

Details are generalised to respect employer and client confidentiality. Numbers shown are the ones I can share.

Building something similar?

Tell me about it. I'm happy to talk architecture, trade-offs and timelines.

Start a conversation