MockArch

Founder / Full Stack Engineer · Public beta

AI-assisted system design interview practice: architecture canvas, automated checks, traffic simulation, and subscriptions (live at mockarch.in).

React 19 · TypeScript · FastAPI · Python · Supabase (Postgres + RLS) · Google Gemini · Razorpay

Open mockarch.inEmail about this

Overview

Owned UI, APIs, AI integration, payments, Supabase RLS, and deployment. Built the product as a multi-tenant SaaS.

React 19TypeScriptFastAPIPythonSupabase (Postgres + RLS)Google GeminiRazorpay

Case study

01 — Situation

Context

System design prep is usually static prompts and generic feedback, which makes it hard to practice repeatedly and improve quickly.

02 — Task

Objective

Build a repeatable practice environment with actionable feedback while keeping user data isolated and usage costs controlled.

03 — Action

Execution

  • 01Built a drag-and-drop architecture canvas with reusable components and persistence.
  • 02Implemented 20+ automated architecture checks (SPOFs, redundancy gaps, queue misconfigurations).
  • 03Added traffic simulation to visualize request flow and bottlenecks.
  • 04Integrated Gemini for feedback with server-side usage limits and cost-aware prompts.
  • 05Shipped subscriptions with Razorpay; enforced multi-tenant isolation using Supabase RLS policies.

04 — Result

Outcomes

  • 01Live product with core flows production-ready and iterating in public beta.
  • 02Clear, scannable feedback loop: checks + simulation + AI review to accelerate practice quality.