The Sentinel: AI-Powered Zero-Touch Insurance for Gig Workers

🚀 From Idea to Execution: Building Zero-Touch Insurance in Phase 2 Introduction

In Phase 1 of Guidewire DEVTrails 2026, we focused on understanding the problem and shaping our idea.

Phase 2 was different.

It was no longer about ideas — it was about making them work in the real world. And that’s where the real challenge began.

🔍 The Problem We’re Solving

Gig delivery workers are the backbone of today’s fast-moving economy. From food delivery to e-commerce, they keep everything running.

But their income is highly unpredictable.

External disruptions like:

Heavy rain Extreme heat Air pollution

can stop them from working — and when that happens, they lose income instantly.

👉 A single day of disruption can mean zero income.

The biggest issue? There is no automatic system to protect them.

⚡ Our Shift in Phase 2

Initially, we made a mistake.

We tried to build everything at once — multiple features, complex logic, and advanced ideas.

Instead of improving our solution, it created confusion and reduced clarity.

So we stepped back and made a key decision:

👉 Focus only on what truly matters.

That shift helped us move from complexity to clarity.

🧩 What We Built

In Phase 2, we focused on building a simple, functional, and automated system.

👤 Worker Onboarding

A clean onboarding flow where users can register, select their platform, and choose their working location.

💰 Weekly Insurance Model

We introduced a flexible weekly subscription model:

₹30 / ₹50 / ₹70 plans Affordable and easy to choose Premium based on risk level 🧠 Dynamic Premium Calculation

Basic AI-driven logic to adjust pricing:

High-risk areas → Higher premium Low-risk areas → Lower premium 🌧 Parametric Triggers

The core of our system.

We defined automatic triggers based on environmental conditions like:

Heavy rain Extreme heat High pollution levels

👉 When thresholds are crossed, the system reacts instantly.

⚙️ Automated Claim System

Unlike traditional insurance:

❌ No forms ❌ No manual claims

✔ Automatic detection ✔ Instant claim generation

💸 Instant Payout Simulation

Once a claim is triggered, the system processes a payout immediately.

Example: 👉 “₹1600 credited”

While simulated, it demonstrates real-world potential.

🏗 System Architecture (Simplified)

Our system is designed with clarity and scalability:

Frontend → User interface for workers Backend → Business logic and claim processing APIs → Weather and environmental data Database → User and policy storage

👉 The focus was simplicity, not complexity.

🧠 Key Learnings 💡 Simplicity Wins

Trying to build everything reduces clarity. A simple system works better.

💡 Automation is Powerful

The best user experience is when the user doesn’t need to act.

💡 Focus on Impact

We shifted from building features to solving real problems.

⚠️ Challenges We Faced Defining accurate trigger thresholds Handling real-time vs mock data Designing a fully automated workflow

Each challenge helped us refine our solution further.

🔮 What’s Next (Phase 3)

In the next phase, we plan to:

Strengthen fraud detection Build advanced dashboards for users and admins Improve payout simulation with payment integrations 🚀 Conclusion

Phase 2 was not just about building a system — it was about building the right system.

We moved from: ❌ Overthinking ➡️ ✔ Clear execution

Our goal is simple: 👉 Zero-touch insurance — where users don’t file claims, the system handles everything.

Because in the real world, protection shouldn’t be complicated.

It should be instant, invisible, and reliable.