The Ai Powered web based Application For Liquidator’s
Author : CA IP SUMIT GUPTA
01 | EXECUTIVE SUMMARY
INSOLIQ is India's first dedicated, AI-assisted web-based Liquidation Process Management System (LPMS) built exclusively for Insolvency Professionals handling company liquidations under the Insolvency and Bankruptcy Code (IBC), 2016. The platform was conceptualised, designed, and deployed entirely using AI-assisted development through prompt engineering.
The project demonstrates how a practising CA and Insolvency Professional can leverage AI tools to build a complete, production-ready SaaS application without traditional software development expertise — using structured Chain-of-Thought prompting as the primary development methodology.
“What started as prompts became a platform. What started as an idea became INSOLIQ.COM”
| 25+ | Features Built | Complete modules | 18 | IBBI Forms | Forms + Records | 3 Days | Days to Build | Concept to Live | Active | Live Users | www.insoliq.com |
02 | THE PROBLEM INSOLIQ SOLVES
Manual Form Generation
IPs manually type IBBI-prescribed forms (B, C, D, E, F, G) and liquidation records for every case, consuming hours of repetitive work.
Missed Statutory Deadlines
Critical IBBI deadlines computed from LCD are tracked manually in spreadsheets, leading to compliance risks and potential NCLT censure.
No Centralised Asset Tracking
Asset valuations, realizations, and sale proceeds are maintained across multiple disconnected Excel files with no single source of truth.
PUFE Transaction Blindspots
Avoidance transaction tracking under Sections 43, 45, 49, 66 IBC lacks a structured system, creating risk in complex liquidations.
Reporting Burden
Quarterly progress reports, final reports, and Section 54 applications require rebuilding the same data from scratch every time.
03 | THE INSOLIQ SOLUTION
INSOLIQ addresses every pain point with a unified, intelligent platform. Built on a modern tech stack (React 18 + TypeScript + Supabase + TanStack Router), the system is accessible from any device via a web browser at www.insoliq.com.
IBBI Statutory Forms
Form B, C, D, E, F, G — auto-populated from case data. Live split-screen preview with DRAFT watermark. PDF download.
12 Liquidation Records
Complete set: Asset Memorandum, Progress Report, Section 52 Notice, E-Auction Notice, Distribution Account (Sec.53 Waterfall), SCC Notice, Disclaimer, Cost Statement, Intimation to Contributories, Final Report (Reg.45), Section 54 Application.
Creditors Management
Add, edit, delete creditors. Bulk upload via Excel template. Export to Excel. Summary bar with total claims and admitted amounts.
Asset Management
Track assets with dual valuer valuations, average computation, realization tracking. Bulk upload. Mark as sold with purchaser details.
PUFE Transactions
Track preferential, undervalued, fraudulent, extortionate transactions under §§43/45/49/66 IBC. Bulk upload. Auditor report upload. Summary notes.
LCD-Based Deadline Tracker
7 statutory milestones auto-computed from Liquidation Commencement Date. Visual timeline with hover tooltips. Compliance alerts.
11-Stage Process Tracker
Metro-style visual timeline of all 11 liquidation stages from Commencement to Dissolution. Progress percentage tracking.
Admin Panel + Licensing
Hidden /admin route for generating and revoking time-limited license keys (INSOLIQ-2026-XXXXXXXX format). Server-side password protection.
Demo/Active Mode System
Unlicensed users get read-only Demo mode. License key activates full write access. Keys managed via Supabase with expiry control.
04 | CHAIN OF THOUGHT PROMPTING — THE DEVELOPMENT METHODOLOGY
The entire INSOLIQ platform was built using Chain-of-Thought (CoT) Prompting — a structured approach where complex tasks are broken into sequential, logical steps. Each prompt built on the output of the previous one, creating a coherent development chain from concept to deployment.
1 FOUNDATION — Define the problem domain
Sample prompt: "Build a Liquidation Process Management System for Indian IPs under IBC 2016 with dark navy + gold UI, Supabase backend, React + TypeScript, shadcn/ui components."
■ Output: App shell, routing, auth, dashboard
2 CORE DATA MODULES — Build primary data entities
Sample prompt: "Add Creditors tab with bulk upload via Excel template, summary bar showing total claims vs admitted amounts, export to Excel using SheetJS."
■ Output: Creditors, Assets, Deadlines, Documents tabs
3 STATUTORY COMPLIANCE — Add IBC-specific features
Sample prompt: "Add LCD Tracker with 7 statutory milestones computed from Liquidation Commencement Date. Show as visual timeline with PA, CSD, SCC, PR, GCS, CL, DA nodes."
■ Output: LCD Tracker, Process Timeline, Stages
4 PUFE MODULE — Specialist avoidance transaction tracking
Sample prompt: "Add PUFE tab for tracking preferential, undervalued, fraudulent, extortionate transactions under §§43/45/49/66 IBC with bulk upload, auditor report upload, and summary notes."
■ Output: Complete PUFE module with all features
5 FORMS GENERATION — Automate statutory form creation
Sample prompt: "Build Forms tab with Form B through G auto-populated from case data, split-screen dialog with live IBBI-style preview, DRAFT watermark, and PDF download."
■ Output: All 6 IBBI forms + 12 Liquidation Records
6 LICENSING & DEPLOY — Commercialise and secure
Sample prompt: "Build hidden /admin route with server-side password protection for generating INSOLIQ-2026-XXXX-XXXX license keys with expiry dates, stored in Supabase licenses table."
■ Output: Admin panel, licensing, live deployment
05 | TECHNOLOGY STACK & AI-ASSISTED DEVELOPMENT
| Layer | Technology | Purpose |
| Frontend | React 18 + TypeScript + Vite | Component-based UI with type safety |
| Routing | TanStack Router | File-based routing with type-safe navigation |
| UI Components | shadcn/ui + Tailwind CSS | Dark navy + gold professional design system |
| Backend/DB | Supabase (PostgreSQL) | Real-time database with Row Level Security |
| State Management | Zustand | License store and global state |
| Forms | React Hook Form + Zod | Validated forms with TypeScript schemas |
| Excel | SheetJS (xlsx) | Bulk upload templates and data export |
| Charts | Recharts | Data visualisation on dashboard |
| Notifications | Sonner | Toast notifications for user feedback |
| AI Development | Claude (Anthropic) | Entire codebase generated via prompting |
| Deployment | Lovable + Custom Domain | Live at www.insoliq.com |
06 | PLAN MODE vs BUILD MODE — AI DEVELOPMENT STRATEGY
A key learning from this project was mastering the two-mode development workflow in Lovable (AI-powered IDE):
| PLAN MODE (0.5 credits) | BUILD MODE (2-5 credits) | |
| What it does | Shows intended file changes without touching code | Actually writes/modifies the codebase |
| When to use | Complex multi-file features, risky changes | Simple fixes, single-component changes |
| Key benefit | Catch wrong assumptions before spending credits | Fast implementation of verified plans |
| Example | Admin panel design (8 files affected) | Timeline date fix (1 component) |
| Credits saved | Prevented ~15 failed builds = ~45 credits | Direct builds on simple tasks |
07 | DATABASE ARCHITECTURE
INSOLIQ uses Supabase (PostgreSQL) with Row Level Security (RLS) policies ensuring each user can only access their own data. The schema was designed iteratively through prompting:
| Table | Key Columns | Purpose |
| cases | case_number, company_name, cin, court, lcd, ip_name, status | Core case registry |
| creditors | case_id, name, type, claim_amount, admitted_amount, status | Claims management |
| assets | case_id, description, category, valuer1_value, valuer2_value, realized_value, status | Asset tracking |
| deadlines | case_id, title, due_date, status, regulation_ref | Compliance deadlines |
| documents | case_id, title, doc_type, file_url, uploaded_at | Document storage |
| forms | case_id, form_code, status, payload (jsonb) | Generated form data |
| pufe_entries | case_id, transaction_date, party_name, transaction_type, amount, status | PUFE tracking |
| case_stages | case_id, stage_name, status, started_at, completed_at | Process stages |
| licenses | key, email, valid_until, issued_at, notes | License management |
| audit_log | case_id, action, entity, entity_id, metadata | Audit trail |
08 | LIVE DEPLOYMENT & METRICS
| Live URL | www.insoliq.com | Deployed on Lovable with custom domain |
| Admin Panel | www.insoliq.com/admin | Hidden route, server-side password protection |
| Day 1 Analytics | 11 visits, 3 countries | India, USA, Italy — organic discovery |
| Device Split | 64% Desktop, 36% Mobile | Responsive design working perfectly |
| Build Time | 3 days (concept to live) | Using AI-assisted development exclusively |
| Credits Used | ~168 Lovable credits | Full production app for ~$50 equivalent |
09 | KEY LEARNINGS FROM THE PROJECT
As a practising IP, domain knowledge made every prompt precise and effective. The AI handled the code; the professional handled the domain requirements.
Breaking the app into sequential modules (Foundation → Data → Compliance → Forms → Licensing) mirrors traditional software architecture. Each prompt was a design decision, not just a request.
Using Plan Mode (0.5 credits) before Build Mode (3-5 credits) prevented multiple expensive failures. Reviewing what files would change before approving a build is equivalent to code review in traditional development.
Vague prompts produce vague results. Specifying exact field names, table structures, color codes, IBC section references, and component patterns in each prompt dramatically improved output quality and reduced iteration cycles.
INSOLIQ is not a prototype — it is a production application handling real legal compliance workflows. This project demonstrates that AI tools have crossed the threshold for building domain-specific professional applications without traditional development teams.
10 | CONCLUSION & FUTURE ROADMAP
INSOLIQ represents a genuine contribution to the Indian insolvency ecosystem. With 3,000+ IBBI-registered Insolvency Professionals in India managing hundreds of active liquidation cases, the market need is real and immediate. The platform is now live, licensed, and ready for commercial deployment.