PropCheck — AI Property Trust Score for India

Problem

Indian property buyers lose lakhs to fraudulent listings on Magicbricks, 99acres, Housing.com, and NoBroker. Fake RERA numbers, recycled stock photos, and inflated pricing slip past buyers because no neutral tool exists to verify a listing in seconds.

Business Impact

Housing.com's own survey says 65% of buyers can't trust listings. Banks and NBFCs spend ₹2,000–5,000 and 3–7 days on property due-diligence per home loan file. PropCheck closes both gaps with one neutral AI engine — free for buyers, API-priced for lenders.

System Approach

• Next.js 14 App Router frontend with Tailwind for the consumer landing, /how-it-works, and /for-lenders pages
• FastAPI 0.115 + Pydantic 2.9 backend exposing /v1/check, /v1/feedback, /healthz endpoints — OpenAPI auto-generated at /docs
• PostgreSQL 16 + SQLAlchemy 2 + Alembic migrations for listing checks, feedback, and the locality price index
• httpx + BeautifulSoup4 + lxml scrape pipeline with Gemma 4 31B (OpenRouter free) as LLM parsing fallback when sites are SPA or rate-limited
• imagehash + Pillow for perceptual-hash reverse-search against a known stock-photo and recycled-listing dataset
• slowapi for IP-level rate limiting; structlog for structured request logs
• Chrome MV3 extension surfaces the Trust Score directly on Magicbricks/99acres/Housing/NoBroker product pages
• Vercel hosts both the Next.js frontend (propcheck.rohitraj.tech) and the FastAPI backend (api.rohitraj.tech) as serverless

Key Decisions & Trade-offs

• Free forever for buyers, paid API for lenders — neutrality is the moat; if we monetized buyers we'd be just another portal
• LLM parsing fallback over heavier headless scrapers — Gemma 4 31B on OpenRouter free tier kept the bundle and the bill at zero
• 8 explicit signals over a black-box ML score — Indian buyers and lenders both want explainable red flags they can act on
• Karnataka RERA first (Bangalore launch May 2026), expand state-by-state — RERA APIs and formats differ per state
• Chrome extension as a wedge — buyers already browse listings on the portals; meet them on the page they're already on

Current Status

Live in production at https://propcheck.rohitraj.tech/. FastAPI backend live at https://api.rohitraj.tech/ with /v1/check + /v1/feedback shipping. LLM parsing fallback verified live (Gemma 4 31B). Chrome MV3 extension built (phase 8). Day 14 calibration on real Magicbricks/99acres URLs revealed SPA + rate-limit walls — calibration deltas in active tuning.

Roadmap

• Tune calibration deltas on real-world URLs from day-14 report
• Harden scraper for SPA + rate-limit walls (Browserbase / residential proxy fallback)
• Expand RERA coverage from Karnataka to Maharashtra, Telangana, and Tamil Nadu
• Ship Chrome extension to the Web Store after MV3 review
• Onboard first lender to the /v1/check API for home-loan diligence

What I'd Improve Next

• Could add a community-sourced scam-listing reporting flow with signed feedback
• Consider WhatsApp delivery of Trust Scores for users who paste links there (currently parked per commit 553c334)

Explore More

• MyFinancial — Personal Financial Advisor — Financial planning in India is fragmented across banks, insurance, and tax documents.
• StellarMIND — Chat-to-SQL with RAG — Business users need to query databases without knowing SQL.
• ClinicAI — WhatsApp AI Clinic Assistant — India has 12 lakh+ small clinics running on phone calls and paper diaries.
• MicroItinerary — AI Travel Planner — Travel apps optimize for proximity and ratings.
• SanatanApp — Hindu Devotional App — Devotional users in India juggle 5+ separate apps for Chalisa, Gita, Aarti, Ramayan, and Mahabharat.
• SynFlow — Enterprise Intelligence Platform — Private deal networks rely on manual introductions and spreadsheets.
• FinBaby (Jama) — Personal Finance Tracker — Indian middle-class families track expenses across UPI apps, bank statements, and paper notebooks.
• RetailOS — Multi-Tenant Retail SaaS — Indian kirana stores and small retailers use paper registers or basic billing software with no inventory tracking, no GST compliance, and no offline support.
• TripHive — Offline-First Collaborative Trip Planner — Group trip planning is fragmented across WhatsApp, Google Docs, Maps, Splitwise, and email.
• ScamRakshak — On-Device AI Scam Detector — Indians lose thousands of crores annually to digital scams via WhatsApp, SMS, and social media.
• PaisaGuard — Family Budget Survival App — Middle-class families worldwide track expenses inconsistently — UPI apps show transactions but don't enforce budgets.
• rohitraj.tech — Engineering work is often invisible.
• Agent Autopsy — Forensic Debugger for Failed AI Agent Runs — Most AI agent failures are silent — the run completes, the status code is green, and the result is still wrong.
• tinyvoice — Fine-Tune a Model in Your Own Voice in an Afternoon — Training your own language model sounds like a PhD job that needs a GPU cluster, so most developers never try.
• snap3d — One Photo In, an Editable 3D Model Out — A photo shows you one side of an object; the other five sides are a guess.
• Agentic OS — Force-Directed Map of a 387-Skill Claude Code Setup — An agent setup grows one plugin at a time until it has hundreds of skills, and nobody can see what it actually contains.
• claude-autodev — Autonomous 8-Stage Dev Pipeline for Claude Code — Agentic coding tools stop at a diff.
• KisanSathi — Six AI Farm Experts, Keyless and Local — Farmer-facing AI tools die at the API-key step, answer in English, and hand back generic advice with no live numbers behind it.
• MarginChef — AI Agent That Finds a Restaurant's Margin Leaks — Restaurants run on 3-5% net margins while food costs are up roughly 35% since 2019, and most owners never see per-dish economics.
• Quorum — Deep-Research Agent Swarm with a Shared GraphRAG Brain — Most AI research tools are one model in a loop.
• RegexForge — Plain English to a Regex You Can Trust — Regex is not hard because the syntax is exotic.
• Skillet — Turn Any Docs Page Into an Installable Claude Code Skill — Agent skills are the fastest way to teach a coding agent a new tool, but writing a good SKILL.
• Ladle — Open-Source Prep Forecasting and Food-Cost Leak Detection — Restaurants throw away 4-10% of the food they buy before it reaches a plate, and most owners find out at month end from a food-cost percentage that moved the wrong way, with no idea which ingredient did it.
• Casita — Design Your Home in 3D in the Browser — Home design tools want an install, a login, or a CAD background.
• VoxelForge — A Voxel Sandbox Engine Built Properly — Voxel sandboxes are the demo everyone builds with a coding agent right now, and almost all of them are a throwaway single HTML file that hitches on chunk generation, ships a texture pack, and cannot be read or extended.
• prompt-ocean — Type a Sea, Watch It Exist — Generative interfaces usually hand a model the wheel and hope.
• HEXAPOD — Inverse Kinematics and Gait Simulator — Hexapod gait and leg IK are usually explained with equations and a video, or hidden inside a robotics library.
• avatar-sync — Real-Time Face and Hand Tracking in One HTML File — Face and hand tracking demos come wrapped in a build step, a server, and usually an API key — which puts a wall in front of anyone who just wants to see what the models actually output before building on them.
• Reliability & Production Readiness — Load testing, observability, and API contracts.
• Open Source Repos — Browse the source code behind these projects.