🚀 Excited to launch my latest project 👉 Zenix!
After months of building bit by bit (and refactoring way too many times 😅), it’s finally ready with that extra layer of polish.
Zenix is an end-to-end platform that simplifies Indian tax prep 👉 no more spreadsheet chaos, no more juggling PDFs.
With this, I might just replace a few CAs (kidding… or am I? 😂).
🎥 Check out the full demo: edited entirely by yours truly.
🧩 Tech Stack Breakdown
Frontend:
React 18 + TypeScript SPA powered by Vite, styled with Chakra UI, Tailwind, and Material UI.
Routing and state flow via React Router, context-based auth guards, toast notifications, Axios, Framer Motion for animations, Lucide/React Icons, and Recharts for analytics visualizations.
Backend:
Built with FastAPI (async routes) handling document ingestion, uploads, tax logic, and analytics.
Uses Pydantic for configuration, centralized logging, and async SQLAlchemy for data access.
Document intelligence routes PDFs/CSVs/XLS/images to processors (Form 16, payslips, or bank statements).
There’s also a Node/Express sidecar with Multer + pdf-parse for auxiliary pipelines.
AI Power:
Landing AI drives extraction.
OpenAI is used for insight generation, it composes regime-specific tax summaries, deductions, and spending recommendations from structured prompts.
Database:
PostgreSQL with async + sync engines, JSON fields, and models covering everything from assessment rollups to categorized transactions.
Auth:
bcrypt via Passlib, JWT-based tokens with strict verification, and multi-level route protection.
(Still expanding on security because in finance, it’s everything.)
💡 Still a lot more to do, but I’m proud of how far it’s come.
Would love to hear your thoughts, suggestions, or even your “CA panic” reactions 😄
#Zenix #TaxAutomation #FastAPI #React #OpenAI #LandingAI #FinTech #IndieDev #Python #TypeScript #Vite #AI #ProjectLaunch #DeveloperJourney #StartupLife