Introduction

Oasis

Parametric wage protection for India’s Q-commerce delivery partners.

When external disruptions hit—extreme heatwaves, government lockdowns, or severe traffic gridlocks—riders lose their ability to complete orders and their daily wage drops sharply. Oasis detects these events via weather and news APIs and provides automated payouts to policyholders without requiring manual claims or waiting periods.

Scope

Oasis strictly covers loss of income associated with external systemic disruptions. It does not provide health, life, accident, or vehicle repair coverage.

---

Executive Summary

Oasis is a Next.js 15 + Supabase application utilizing a Parametric Insurance model. Unlike traditional indemnity insurance where a user must prove their loss, parametric insurance pays out automatically when predefined data parameters (e.g., Temperature > 43°C) are met.

This completely eliminates the overhead of claims adjusters, fraud investigators, and processing delays, instantly crediting the rider’s wallet the moment a disruption event is validated.

The Problem Statement

Companies like Zepto and Blinkit operate on incredibly tight 10-minute delivery SLAs across hyper-local zones. A rider’s income is directly tied to their volume of completed trips.

However, riders operate in highly volatile physical environments.

1. Unpredictable Weather: Heatwaves force riders to log off for safety, while sudden heavy rains flood roads, slowing completion times.
2. Civic Disruptions: Strike actions, VIP movements, or localized curfews (Section 144) can forcefully shut down entire delivery zones for hours.
3. The Proof Burden: When these events happen, riders lose income. Attempting to claim compensation from traditional insurers requires filing paperwork, proving the loss, and waiting weeks-a process incompatible with a gig worker’s daily cashflow needs.

Extreme Heat

A rider logs off because it is physically unsafe to ride. Oasis detects the heatwave via weather APIs and auto-credits the protected wage.

Zone Lockdown

A curfew blocks delivery operations. News APIs and LLMs verify the police order and trigger instant payouts for riders within the geofence.

Heavy Rain

Localized flooding halts area movement. Precipitation thresholds are met, and policyholders immediately receive automated payouts.

The Oasis Solution

By removing the human element from the claims process, Oasis creates a highly scalable, zero-friction protection net.

Technical Implementation Workflow

1. Continuous Ingestion: A Vercel Cron (runAdjudicator) polls active delivery zones every 15 minutes.
2. Data Validation: The adjudicator queries Tomorrow.io (weather), WAQI/Open-Meteo (AQI), TomTom (traffic), and NewsData.io (civic issues). If a headline suggests a disruption, an OpenRouter Vision/Text LLM verifies the severity and extracts the affected geolocations.
3. Parametric Trigger: If the parsed data crosses a predefined threshold (e.g., Severity >= 6/10), the system generates a live_disruption_event.
4. Instant Adjudication: The system finds all riders with active policies inside that event’s geofence, runs an 11-step automated fraud detection pipeline, and immediately inserts parametric_claims.
5. Real-time Settlement: Supabase Realtime broadcasts the database insertion directly to the rider’s dashboard, updating their wallet balance instantly without a page reload.

How It Works

flowchart LR
    O[Onboarding]
    S[Weekly coverage]
    P[Payment]
    E[Automation engine]
    H[Runs every 15 minutes]
    Q{Disruption detected?}
    F[Safety & fraud checks]
    A[Claim created automatically]
    V[Quick GPS confirmation]
    W[Wallet updates instantly]

    O --> S
    S --> P
    P --> E
    E --> H
    H --> Q
    Q -->|No| E
    Q -->|Yes| F
    F --> A
    A --> V
    V --> W

    classDef user fill:#BFE3FF,stroke:#2B3A55,stroke-width:1.5px,color:#111;
    classDef system fill:#FFF4DD,stroke:#Caa24A,stroke-width:1.5px,color:#111;
    classDef decision fill:#EAEAEA,stroke:#5A5A5A,stroke-width:1.5px,color:#111;
    classDef success fill:#BFF0C4,stroke:#2F6F3E,stroke-width:1.5px,color:#111;

    class O,S,P user;
    class E,H,F,A,V system;
    class Q decision;
    class W success;

1. Onboard

Platform, zone, gov ID, face verification.

2. Subscribe

Weekly coverage (₹49–₹199), dynamic by zone risk.

3. Monitor

15-min cron polls weather, AQI, traffic, news.

4. Payout

Threshold hit → adjudicator creates claims → Supabase Realtime updates wallet.

Dynamic Premium Plans

Coverage pricing is generated by a deterministic, actuarial constraint engine. It calculates a unified raw Risk Score (0-1) across 5 pillars:

1. Zone History (35%)
2. Weather/AQI Forecasts (25%)
3. Income Exposure (15%)
4. Social Disruption Frequency (10%)
5. Rider Claim Behavior (15%)

Strict Solvency & Option B

The system mechanically enforces a 40% profit margin on calculated expected losses. During extreme, guaranteed-loss cyclonic weather (A “Black Swan”), the system overrides raw math and invokes “Growth Mode” (Option B). It caps expected exposure to 1 claim/week and forcibly floors absolute payouts at 1.5x the highest premium to heavily subsidize rider satisfaction, effortlessly absorbed by the 40% margin built up during regular weeks.

Prices mathematically float, tightly clamped between ₹49 – ₹199/week. The engine generates 3 strict options:

Basic Coverage

0.7x Base Premium

• Payout: Base limit scaled down
• Limit: 1 Claim / week

Standard Coverage

1.0x Base Premium

• Payout: Nominal Base Limit
• Limit: 2 Claims / week

Premium Coverage

1.3x Base Premium

• Payout: 1.5x boosted limit
• Limit: 3 Claims / week

Stack

Framework

Next.js 15

Database & Auth

Supabase

AI / LLM

OpenRouter

Data Triggers

Tomorrow.io, WAQI, TomTom, NewsData.io

Payments

Razorpay · Standard Checkout (UPI, cards, INR)

Hosting

Vercel