My Journey Through the AI Agents Intensive, Building an AI Personal Safety & Emergency Assistant

MUKUL TIWARI — Fri, 05 Dec 2025 02:44:38 +0000

How I Built an AI Personal Safety & Emergency Assistant Using Multi-Agent Systems

The 5-Day Google & Kaggle AI Agents Intensive Course was one of the most transformative learning experiences I’ve had. I came into the course excited but unsure about how complex multi-agent systems really work. By the end, I built a complete AI Personal Safety & Emergency Assistant—a multi-agent emergency detection system that analyzes risk, triggers alerts, and assists users during dangerous situations.

This post is a reflection on what I learned, how the course changed my understanding of agentic AI, and how I built my capstone project.

What I Learned From the 5-Day Intensive

Before this course, my idea of “agents” was very basic. But the training revealed how powerful AI agents are when combined:

Day 1 — The Fundamentals

I learned the foundations: reasoning loops, agent instructions, routing, and agent orchestration.

Day 2 — Tools

This day taught me how agents use tools to extend their abilities beyond text — a key concept I used later to build simulated SMS, email, and call alerts.

Day 3 — Memory

This unlocked a turning point for me. Agents that remember past interactions behave more intelligently.

I implemented risk trend detection where repeated danger messages trigger escalation.

Day 4 — Evaluation & Observability

This taught me how to test, track, and debug agent behavior — essential for safety systems.

Day 5 — Agent-to-Agent Communication

I learned how agents can collaborate to form a full workflow.

This is what allowed me to design a multi-agent pipeline for emergency response.

My Capstone Project: AI Personal Safety & Emergency Assistant

Inspired by real-world safety challenges, I built a system that acts when people can’t call for help themselves.

Problem

Millions face emergencies where they are unable to call or message family or police.

Seconds matter — and AI can react faster than humans.

Why Agents?

Agents fit this problem because they can:

Detect danger from text inputs
Decide the correct action
Trigger emergency-like responses
Guide users step-by-step
Escalate automatically when needed

Traditional chatbots cannot do this. Agents can.

Architecture I Built

My system includes three cooperating agents:

1. Risk Detector Agent

Classifies messages as:

SAFE
EMERGENCY

2. Action Planner Agent

Decides what to do:

reassure
ask more details
escalate to emergency mode

3. Responder Agent

Provides urgent step-by-step instructions in critical situations.

Memory Module

Tracks:

previous messages
previous risk levels
escalation patterns

Tool Simulation

I built safe simulated tools:

send_sms_alert()
send_email_alert()
send_call_alert()

Gemini Integration (Mock Model)

Demonstrates:

danger classification
emergency message generation

All of this was done inside a Kaggle Notebook.

What I Tested

I evaluated my agents across scenarios:

Clear emergency (“I am bleeding, please help!”)
Safe message (“I reached home safely.”)
Ambiguous risk
Escalating danger (“Someone is following me → I am in danger → He is attacking me”)

The system behaved consistently, escalated correctly, and triggered alerts responsibly.

What This Project Taught Me

1. Multi-agent systems are extremely powerful

Combining simple agents created a system far more intelligent than a single model.

2. Memory changes everything

The moment agents remember context, their decisions become smarter.

3. Tools transform agents into action-takers

Even simulated tools felt like building the foundation of a real safety product.

4. Clear instructions matter more than code

The course proved that well-written agent instructions are as important as model power.

If I Had More Time

I would add:

Voice-based danger detection
GPS-based location alerts
Mobile app interface
Real API integrations (Twilio, WhatsApp)
Deployment on Cloud Run / Agent Engine

Final Thoughts

The Google × Kaggle AI Agents Intensive didn't just teach me agents —

It taught me how to build impactful, real-world AI systems.

My capstone project, AI Personal Safety & Emergency Assistant, is only the beginning.

And this course has opened the door to a new world of possibilities.

Thank you Google, Kaggle, and the entire AI Agents community.

DEV Community: MUKUL TIWARI