Computer Science & Engineering (Data Science) Undergraduate • Technical Product Management Aspirant • Systems Builder
Christ College of Engineering (APJ Abdul Kalam Technological University) • Kerala, India

"Building technology. Thinking in products."

I am a Computer Science & Engineering (Data Science) undergraduate developing toward Technical Product Management (Tech PM), supported by hands-on full-stack engineering, applied machine learning, and project leadership.

I specialize in bridging the gap between business objectives, user workflows, and technical system architecture:

• Product Strategy & Problem Scoping: Deconstructing complex, fragmented operational pain points into prioritized product roadmaps and deterministic user journeys.
• Engineering Execution & System Integrity: Architecting resilient full-stack platforms with strict mathematical guarantees (e.g. idempotent billing engines, client-side offline OCR, relational star schemas).
• Data-Driven Decision Making: Leveraging analytical pipelines, telemetry modeling, and statistical forecasting to measure feature impact and eliminate commercial ambiguity.
• Cross-Functional Collaboration: Leading multi-disciplinary hackathon teams from initial problem framing through technical specification, implementation, QA, and executive stakeholder demo.

Role: Team Leader, Technical Product Lead & Full-Stack Architect • Hackathon: Odoo × Make-A-Ton 9.0 Kochi Hackathon (2026)
Product Value: Eliminating SaaS revenue leakage and invoice disputes through automated billing operations and predictive cash forecasting.

• The Problem: Subscription-based SaaS companies face severe revenue leakage: manual invoicing errors, missed renewal dates, untracked overdue payments, and double-billing disputes caused by spreadsheet-driven billing operations.
• Target User: Finance operations leads, CFOs, and billing administrators managing high-volume recurring customer accounts.
• Product Solution: A unified, production-grade recurring revenue management platform managing the end-to-end customer-to-cash lifecycle: dynamic product/plan catalogs, quotations, automated batch billing, GST compliance, partial payment reconciliation, and 30-60-90 day forward-looking predictive cash forecasting.
• Technical & Architectural Execution:
  • Zero Double-Billing Guarantee: Database-level UniqueConstraint(subscription, period_start) preventing duplicate invoice generation across recurring cycles.
  • Financial Precision: Integer-paise database storage (100 paise = ₹1.00) eliminating floating-point rounding inaccuracies.
  • Predictive Cash Forecasting: Mathematically rigorous 30-60-90 day projection engine separating Contractual Expected Billings from Expected Cash Collections using historical collection efficiency factors.
  • Instant Bharat UPI Dynamic QR: Pure JavaScript Galois Field $GF(2^8)$ QR synthesizer (qr.js) generating standard NPCI vector deep links (upi://pay?...) with instant settlement writes to backend.
  • Interactive Time-Travel Sandbox: Temporal simulation engine executing batch billing runs up to +90 days into the future.
• Verified Evidence: 15 / 15 automated backend tests passing (100%), 8 / 8 live E2E lifecycle stages verified, 0 console errors.
• Tech Stack: Python • Django REST Framework • React 18 • Vite • PostgreSQL / SQLite • Tailwind CSS • Recharts

graph LR
    Sub[Active Subscription] -->|Batch Schedule / Cron| Billing[Idempotent Billing Runner]
    Billing -->|UniqueConstraint Check| Inv[Tax Invoice INV/2026]
    Inv -->|Itemized GST Split| Tax[CGST+SGST / IGST Engine]
    Inv -->|Galois Field QR Generator| UPI[Bharat UPI Dynamic Terminal]
    UPI -->|Instant Settlement / Partial| Pay[Payment Reconciliation]
    Pay -->|Real-time Rollup| Dash[30-60-90d Revenue & Cash Forecast]

Role: Technical Product Architect & AI Lead • Event: AIBoomi Startup Weekend / SIH 2026
Product Value: Transforming slow, manual regulatory label audits into instant, client-side statutory screening verdicts with zero cloud privacy risks.

• The Problem: Enforcement officers and compliance inspectors spend hours manually inspecting tiny packaged commodity labels against Indian Legal Metrology regulations, battling reflective packaging glare, crinkled surfaces, and tiny fonts.
• Target User: Legal metrology field inspectors, consumer protection authorities, and FMCG packaging QA teams.
• Product Solution: An offline-first Progressive Web Application with client-side OCR and deterministic statutory rule evaluation, providing immediate PASS or REVIEW legal screening verdicts.
• Technical & Architectural Execution:
  • 100% Client-Side Privacy: Web Workers execute multi-pass image binarization and Tesseract.js OCR locally in browser memory without sending private photos over external networks.
  • Statutory Gazette Rule Engine: Deterministic rule parser cross-referencing statutory Gazette amendments (G.S.R. 202(E), G.S.R. 779(E)) for MRP, Net Quantity, Manufacturer details, and Consumer Care contacts.
  • Offline Sync: Automatic IndexedDB local queue synchronization with FastAPI persistence server upon network reconnection.
• Verified Evidence: 198 frontend Vitest unit/integration tests, 61 backend pytest tests, production deployment live on Vercel.
• Tech Stack: TypeScript • React • Vite • Tesseract.js • Web Workers • FastAPI • PWA • IndexedDB
• 🌐 Live Application: https://lmcc-web-ten.vercel.app

flowchart LR
    Camera[Label Capture] --> Worker[Web Worker Tesseract.js OCR]
    Worker --> Engine[Rule 6 Legal Metrology Engine]
    Engine --> Verdict{Statutory Verdict}
    Verdict -->|PASS / REVIEW| LocalDB[(IndexedDB Offline Queue)]
    LocalDB -->|Auto-Sync| API[FastAPI Persistence Server]

Role: Full-Stack & Automation Product Lead
Product Value: Automating 15+ hours weekly of manual visual asset creation, copywriting, and board syndication for digital commerce.

• The Problem: Skincare affiliate marketers spend hours manually designing vertical pins, writing SEO descriptions, and scheduling board posts across platforms.
• Target User: E-commerce brand operators, affiliate content creators, and growth marketing teams.
• Product Solution: End-to-end automated SaaS content engine transforming luxury skincare product URLs into publication-ready 9:16 vertical graphics, SEO botanical copy, and automated scheduled board syndications.
• Technical & Architectural Execution: Modern SSR/CSR architecture on TanStack Start (React 19) orchestrating decoupled n8n asynchronous webhook queues, OpenAI GPT-4o copy generation, Supabase RLS persistence, and Pinterest API v5 syndication.
• Tech Stack: TypeScript • TanStack Start • React 19 • Supabase (PostgreSQL) • OpenAI GPT-4o • n8n • Pinterest API v5
• 🌐 Live Platform: https://verse-automator.pages.dev

Role: Analytics & Data Product Lead
Product Value: Converting fragmented retail transactional logs into actionable customer retention and payment trajectory intelligence.

• The Problem: Retail organizations struggle to extract actionable operational insights and customer churn patterns from fragmented operational transactional tables.
• Target User: Retail operations executives, merchandising leads, and store managers.
• Product Solution: An end-to-end data analytics suite modeling ₹11,06,505 in sales volume across 320 retail transactions over 18 months, evaluating payment method dominance (UPI vs. Cards) and product category trajectories.
• Technical & Architectural Execution: Normalized Star Schema dimensional architecture linking Fact Orders to Customer, Employee, and Product dimensions, coupled with Pandas data cleaning and interactive Power BI DAX executive dashboards.
• Tech Stack: Python • Pandas • Jupyter Notebook • Power BI Desktop (DAX) • Relational SQL

Role: Backend & Product Systems Lead
Product Value: Creating an auditable, transparent bridge between citizen environmental incident reports and municipal action dispatch.

• The Problem: Citizens lack a transparent, auditable channel to report environmental violations (illegal dumping, industrial discharge) and monitor municipal enforcement.
• Target User: Civic-minded citizens and municipal environmental protection officers.
• Product Solution: Full-stack civic platform featuring geo-tagged incident reporting, real-time status auditing, and administrative dispatch workflows.
• Tech Stack: Java • Spring Boot • PostgreSQL • REST API • React • Tailwind CSS
• 🌐 Live Demo: https://frontend-nine-woad-g8xvq6ys3s.vercel.app

Role: Fintech Product & Full-Stack Lead
Product Value: Protecting MSME cash flow through deterministic invoice aging and automated dispute communications.

• The Problem: B2B MSMEs suffer crippling cash flow delays due to disputed invoices, untracked delivery acknowledgments, and lack of formal statutory demand notices.
• Target User: Small business owners and receivables accounting clerks.
• Product Solution: Receivables copilot enforcing deterministic aging calculations (Current, 1–30d, 31–60d, 60d+), MSME Samadhaan compounding interest calculations, and AI dispute resolution draft letters.
• Tech Stack: FastAPI (Python 3.11) • React • Pydantic • Render • Vercel • Pytest

• 🎓 Degree: B.Tech in Computer Science & Engineering (Data Science)
  Christ College of Engineering, Irinjalakuda (APJ Abdul Kalam Technological University) • 3rd Year
• 🏆 Odoo × Make-A-Ton 9.0 Kochi Hackathon (2026): Team Leader & Technical Product Lead for SubSync — managed sprint planning, architecture, and live E2E delivery.
• 🏆 AIBoomi Startup Weekend Kochi: Team Lead & Technical Architect for JAMH X / LMCC — offline-first statutory compliance screening product.
• 🏆 Smart India Hackathon (SIH 2026): Lead developer for client-side regulatory automation and document verification solutions.
• 💼 Industry Experience: Project Management & Automation Intern — client coordination, Zoho workflow design, and operational process syndication.

"Building resilient architectures that bridge engineering depth, user workflows, and business outcomes."