DEV Community: Omkar Ghate

What Is Agentic AI? The Beginner's Guide to AI Agents in 2026

Omkar Ghate — Thu, 29 Oct 2026 11:48:55 +0000

What Is Agentic AI? The Beginner's Guide to AI Agents in 2026

You've heard the term "AI agents" everywhere in 2026 — but what does it actually mean? This guide breaks down agentic AI in plain language, explains how it's different from regular AI chatbots, and shows you real examples of AI agents working in Indian businesses today.

The Simple Definition

Agentic AI is an AI system that can take actions on its own to complete a goal — not just answer questions.

Regular AI (like a basic chatbot): You ask, it answers. Done.

Agentic AI: You give it a goal, and it figures out the steps, takes actions, uses tools, and completes the task — even across multiple systems.

AI Chatbot vs. AI Agent: What's the Difference?

	AI Chatbot	AI Agent
What it does	Answers questions	Takes actions
Memory	Usually none	Can remember context
Tools	None	Can use APIs, databases, browsers
Autonomy	Zero	High
Example	"What is your price?" → Answers	"Book a demo for tomorrow" → Actually books it

A Real Example: The Difference in Action

Scenario: A customer calls a real estate company wanting to book a site visit.

Basic chatbot:

Customer: "I want to book a site visit"

Bot: "Please call us at 98XXXXXXXX to book a site visit"

Agentic AI:

Customer: "I want to book a site visit for your Pune project"

Agent: "Of course! Are you available this Saturday or Sunday?"

Customer: "Saturday works"

Agent: "Great! What time — morning or afternoon?"

Customer: "11am"

Agent: [Checks calendar availability in real-time]

Agent: "I've booked your visit for Saturday 11am. You'll receive a confirmation on your WhatsApp at XXXXXXXXXX. Our team member Rahul will meet you at the site. Can I help with anything else?"

The agentic system actually booked the appointment, checked the calendar, and sent a WhatsApp confirmation — all automatically.

The 5 Core Capabilities of AI Agents

1. Reasoning

Agents can think through multi-step problems. Instead of matching keywords, they understand intent and figure out the best path to the goal.

2. Tool Use

Agents can use external tools:

Databases (look up customer info)
APIs (send WhatsApp, check inventory)
Calendars (check and book availability)
Web browsers (search for information)
Email systems (send confirmation emails)

3. Memory

Agents can remember context within a conversation — and increasingly, across multiple conversations. They know who you are, what you ordered last time, and what you discussed before.

4. Planning

Given a complex goal, agents break it into sub-tasks and complete them in order. "Book a demo, send a confirmation, add to CRM, alert the sales rep" — all from one customer instruction.

5. Self-Correction

When something goes wrong, good AI agents detect the error and try a different approach rather than failing or giving a wrong answer.

Types of AI Agents in 2026

Voice Agents

AI that communicates via phone calls. These are handling millions of customer service, sales, and support calls daily. Platforms: Bolna AI, Vapi AI, Retell AI.

Workflow Automation Agents

AI that runs business processes automatically. Connect your email, CRM, calendar, and databases, and the agent manages the entire workflow. Platforms: n8n, Make.com.

Research Agents

AI that browses the web, reads documents, and compiles research reports. Give it a topic and come back to a complete briefing.

Customer Service Agents

AI that handles customer queries across WhatsApp, email, chat, and phone simultaneously. These never sleep and scale infinitely.

Sales Agents

AI that qualifies leads, follows up automatically, and schedules meetings with your sales team only when a lead is ready to buy.

Real Agentic AI Examples in Indian Businesses

TVS Motors — Lead Qualification Agent

An AI voice agent calls dealer network leads, qualifies them with 6 key questions, and routes hot leads directly to the regional sales manager. Cold leads receive automated follow-up messages.

Result: Sales team talks only to qualified leads. No time wasted on bad leads.

Hubtown Real Estate — Site Visit Booking

An AI agent handles WhatsApp inquiries about properties, answers detailed questions from the knowledge base, qualifies interest level, and books site visits on the developer's calendar.

Result: Site visits increased 35% without additional staff.

Finideas — Investor Onboarding

An AI agent guides new mutual fund investors through KYC completion, explains fund options based on risk profile, and escalates complex queries to human advisors.

Result: Onboarding time reduced from 3 days to 4 hours.

How to Start Using Agentic AI in Your Business

Start Small

Don't try to automate everything at once. Pick one repetitive process:

Appointment booking
FAQ answering
Lead qualification
Order status updates

Choose the Right Platform

Voice agents: Bolna AI, Vapi AI
WhatsApp agents: Interakt, Wati, or custom via WhatsApp API
Workflow automation: n8n, Make.com
Customer support: Freshdesk AI, Zendesk AI

Measure the Results

Track:

How many interactions the agent handled
Customer satisfaction scores
Time saved for your team
Revenue impact (more leads converted, faster response)

The Future of Agentic AI

We're moving from "AI that answers questions" to "AI that runs businesses." By 2027-2028, most routine business operations will have an AI agent handling them — scheduling, invoicing, customer service, inventory management, and more.

The businesses that understand and deploy agentic AI now will be significantly ahead of those that wait.

The question isn't whether AI agents will transform your business. It's whether you'll be leading that transformation or catching up to it.

Omkar Vijay Ghate is an Agentic AI Architect and the creator of OmiLabs. He has built 16+ live AI agents for Indian businesses across automotive, real estate, finance, and retail. Follow on YouTube @omilabs.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

Voice Cloning for Business in India: Create Your Brand's AI Voice

Omkar Ghate — Thu, 29 Oct 2026 11:47:54 +0000

Voice Cloning for Business in India: Create Your Brand's AI Voice

Imagine having a consistent, branded AI voice that represents your company across every customer interaction — phone calls, IVR systems, video content, and chatbots. In 2026, this is no longer science fiction. Voice cloning technology has made it practical and affordable for Indian businesses.

What Is Voice Cloning?

Voice cloning is the process of training an AI model on a sample of a human voice to create a digital replica that can speak any text in that person's voice. The result is a synthetic voice that sounds nearly identical to the original speaker.

Modern voice cloning requires as little as 30 seconds of audio — though 5-10 minutes produces significantly better results.

Why Indian Businesses Need a Branded AI Voice

1. Consistency at Scale

When you have 10 customer service agents, each sounds different. A cloned AI voice sounds identical every time — same tone, same warmth, same brand personality.

2. Regional Authenticity

Clone a voice that speaks with a Mumbai, Delhi, or Chennai accent to connect better with local customers.

3. Cost Efficiency

Professional voice-over recordings cost ₹5,000–₹50,000 per project. Once you clone a voice, generating new audio content costs almost nothing.

4. Instant Content Creation

Need a new IVR message? New product announcement? Updated terms recording? Generate it in seconds with your cloned voice.

How to Clone a Voice for Your Business

Step 1: Choose Your Voice Actor

This is the most important decision. Options:

Clone yourself — authentic, personal, great for solopreneurs
Hire a voice actor — professional quality, ensure you have rights
Clone an employee — works well for call center agents

Legal note: Always get written consent from the person whose voice you're cloning. This is both ethical and increasingly a legal requirement.

Step 2: Record Quality Audio

Technical requirements:

Sample rate: 44.1 kHz or higher
Format: WAV or high-quality MP3
Duration: Minimum 30 seconds, ideally 5+ minutes
Environment: Quiet room, no background noise

What to record:

Various sentence types (questions, statements, exclamations)
Different emotional tones (friendly, professional, empathetic)
Common phrases your agent will use
Numbers and technical terms your business uses

Step 3: Upload to ElevenLabs

ElevenLabs is currently the best platform for Indian language voice cloning:

Go to elevenlabs.io
Click "Add Voice" → "Clone Voice"
Upload your audio samples
Name your voice and add description
Click "Add Voice" to start training
Testing takes about 30 seconds

Step 4: Test and Refine

Generate test audio with your cloned voice. Listen carefully for:

Pronunciation accuracy (especially for Indian names, places)
Emotional range
Naturalness in long sentences
Handling of numbers and special characters

Step 5: Deploy

Your cloned voice can now be used via:

ElevenLabs API — integrate into any application
Bolna AI — use as your voice agent's voice
Vapi AI — deploy as a customer service agent
Direct API — generate audio files for marketing

Use Cases for Voice Cloning in India

IVR Systems

Replace robotic IVR recordings with a warm, branded voice. "Welcome to [Company Name]. Mera naam Priya hai. Kaise help kar sakti hun aapki?"

Customer Service Agents

Build voice AI agents that sound like a real team member. Customers can't tell it's AI — they just know the experience is good.

Marketing Videos

Generate voiceovers for YouTube videos, ads, and explainer content in minutes rather than scheduling studio sessions.

Audiobooks and Podcasts

Indian authors can now create audiobooks in their own voice at scale — even books they didn't write in audio format initially.

Training Content

Create training videos and e-learning content with consistent narration without recurring recording costs.

Top Voice Cloning Platforms for India

Platform	Hindi Support	Pricing	Best For
ElevenLabs	Excellent	$5/month starter	Quality voice agents
Resemble AI	Good	$0.006/sec	High-volume generation
Play.ht	Good	$39/month	Podcast/content
Murf AI	Basic	₹1,999/month	Marketing content

Ethical Considerations

Voice cloning is powerful and comes with responsibility:

Always get consent from the person being cloned
Disclose when AI voice is used in customer interactions if required by law
Don't clone celebrities without permission
Watermark your cloned voice where possible
Monitor for misuse — revoke access if cloned voice is misused

India's IT rules and emerging AI regulations are increasingly covering synthetic media. Stay updated on compliance requirements.

My Setup for Client Projects

For the 16+ voice agents I've built for Indian businesses, my standard stack is:

Voice cloning: ElevenLabs (instant voice clone)
Agent platform: Bolna AI (handles the conversation logic)
Integration: Webhook to client's CRM
Monitoring: Call recordings reviewed weekly

This setup delivers a production-ready AI voice agent in 3-5 days at a fraction of hiring costs.

Omkar Vijay Ghate is an Agentic AI Architect based in Mumbai, specializing in voice AI for Indian businesses. Follow for more on YouTube @omilabs or connect on LinkedIn.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

Multilingual Voice AI in India: Building Agents That Speak Hindi, English & More

Omkar Ghate — Thu, 29 Oct 2026 11:42:51 +0000

Multilingual Voice AI in India: Building Agents That Speak Hindi, English & More

India is a linguistically diverse nation with 22 scheduled languages and hundreds of dialects. For AI voice agents to truly serve Indian customers, they must go beyond English — and in 2026, this is finally becoming a reality.

Why Multilingual AI Matters for Indian Businesses

The numbers tell the story:

Only 10% of Indians are comfortable conversing in English
55% prefer Hindi for business conversations
Regional languages (Tamil, Telugu, Marathi, Bengali) dominate Tier 2 and Tier 3 cities
Code-switching (mixing two languages mid-conversation) is the natural speaking style of most urban Indians

If your AI voice agent only speaks English, you're leaving 90% of your potential customers underserved.

The Challenge of Building Multilingual AI for India

1. Hindi Is Not Simple

Hindi has formal and informal registers, regional variations (Bhojpuri-influenced, Rajasthani-influenced, Delhi-style), and complex honorific systems. An AI trained on standard textbook Hindi may sound robotic to actual Hindi speakers.

2. Code-Switching Is Normal

Most urban Indians naturally switch between Hindi and English mid-sentence:

"Mujhe ek appointment book karni hai tomorrow"
"Sir, kya aap Monday ko available hain?"
"The product ka price kya hai?"

Your AI agent must understand and respond naturally to this mixed-language style.

3. Accent Recognition

Indian English accents vary dramatically between regions. An AI that doesn't recognize Tamilian English, Punjabi English, or Gujarati English will constantly misunderstand customers.

Top Platforms for Multilingual Indian Voice AI

Bolna AI

Bolna has built specific support for Indian languages. Their models are trained on actual Indian conversational data, making them far more natural than generic models.

Supported languages: Hindi, English, Hinglish, Tamil, Telugu, Kannada, Bengali, Marathi

Key feature: Built-in code-switching — the agent can seamlessly handle sentences that mix Hindi and English.

ElevenLabs for Indian Voices

ElevenLabs offers high-quality Indian voice options for:

Hindi (Male/Female)
Indian English (Mumbai, Delhi, South Indian accents)
Custom voice cloning

You can clone a real human voice and use it for your AI agent, giving it an authentic Indian sound.

Sarvam AI

Sarvam AI is an Indian startup building AI specifically for Indian languages. Their models are trained on Indic language data and handle regional variations better than global models.

How to Build a Hinglish AI Agent

Hinglish (Hindi + English code-switching) is the dominant communication style for urban Indian customers between ages 18-45. Here's how to build a natural Hinglish agent:

Step 1: Write a Hinglish System Prompt

Tu ek friendly customer service agent hai "TechIndia" ke liye.
Tera naam Priya hai. Tu Hinglish mein baat karta hai — 
naturally Hindi aur English mix kar ke. Formal mat bana, 
friendly aur helpful rah. Agar koi Hindi mein baat kare, 
Hindi mein reply kar. Agar English mein, toh English mein.

Step 2: Train on Indian Examples

Add example conversations to your agent's knowledge base using real Hinglish dialogue:

Customer: "Mera order kab aayega?"

Agent: "Sir, aapka order 2-3 din mein deliver ho jayega. Kya aapko tracking update chahiye?"

Step 3: Handle Language Switching

Configure your agent to detect the customer's language preference in the first 2 sentences and match it for the rest of the conversation.

Real-World Use Cases

1. TVS Motors — Lead Qualification

I built a Hindi/English voice agent for TVS Motors that qualifies dealer leads. The agent:

Opens in Hinglish
Detects if the dealer prefers Hindi or English
Conducts a 3-minute qualification conversation
Books a callback if the lead qualifies

Result: 40% more leads qualified per day vs. manual calling

2. Real Estate in Tier 2 Cities

For a Pune real estate developer, I built an agent that speaks:

Marathi for local customers
Hindi for North Indian buyers
English for NRI inquiries

The same system handles all three language groups automatically.

Tips for Better Multilingual AI

1. Use Indian voice actors for cloning
Clone voices of actual Indian speakers, not American or British voices tweaked with an accent.

2. Test with real Indian users
Have 10 actual target customers test your agent before going live. They will catch unnatural phrases that you might miss.

3. Build language detection into the flow
Start every conversation with a language detection step. Ask: "Kya aap Hindi mein baat karna chahenge?" or let the customer's first message determine the language.

4. Handle silence and interruptions
Indian callers often interrupt, add "haan haan" confirmations, or go silent. Configure your agent to handle these naturally.

The Future of Indian Language AI

2026 is the tipping point for Indian language AI. With models from Sarvam AI, Krutrim, and global players adding Indian language support, we're moving toward truly regional AI agents.

The businesses that invest in multilingual AI now will have a significant advantage in reaching Bharat — the 900 million Indians who don't speak English.

Omkar Vijay Ghate is an Agentic AI Architect based in Mumbai specializing in voice AI for Indian businesses. Connect on LinkedIn or subscribe to OmiLabs on YouTube.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

How to Build an AI Agent Without Writing Code (2026 Guide)

Omkar Ghate — Thu, 29 Oct 2026 11:41:50 +0000

How to Build an AI Agent Without Writing Code (2026 Guide)

You don't need to be a developer to build powerful AI agents in 2026. No-code platforms have made it possible for entrepreneurs, marketers, and business owners to deploy their own conversational AI systems — without writing a single line of code.

What Is a No-Code AI Agent?

A no-code AI agent is a conversational AI system built using visual drag-and-drop tools, pre-built templates, and simple configuration interfaces. Instead of writing Python or JavaScript, you connect blocks, set conditions, and define conversation flows using a visual interface.

Top No-Code AI Agent Platforms in 2026

1. Bolna AI

Bolna is one of the best platforms for building voice AI agents without code. You can create a complete AI phone agent in under 30 minutes.

What you can build:

Outbound sales calling agents
Appointment booking bots
Customer service voice agents
Lead qualification systems

2. Vapi AI

Vapi is another powerful voice AI platform with a clean dashboard. It supports multiple LLMs (GPT-4o, Claude, Gemini) and integrates easily with CRMs.

3. n8n (Visual Automation)

n8n lets you build AI workflows visually. You can connect AI models to databases, APIs, email systems, and more — all without code.

4. Make.com

Make (formerly Integromat) is excellent for connecting AI services to your existing business tools like Google Sheets, WhatsApp, Slack, and CRMs.

Step-by-Step: Build The Human Pattern on Bolna

Step 1: Create Your Bolna Account

Go to bolna.ai and sign up for a free account. You get free testing credits to get started.

Step 2: Create a New Agent

Click "Create Agent" and choose your template:

Outbound Sales
Customer Support
Appointment Booking
Custom (blank canvas)

Step 3: Write Your System Prompt

The system prompt defines your agent's personality, goals, and behavior. Keep it clear and specific:

You are Priya, a friendly customer service agent for [Business Name]. 
Your goal is to answer questions about our products, take orders, 
and schedule appointments. Always be polite and professional.
If you cannot help, transfer to a human agent.

Step 4: Choose Your Voice

Select from 100+ AI voices in different languages and accents. For Indian businesses, Hindi and Hinglish voices work best for local customers.

Step 5: Connect Your Tools

Add integrations:

CRM: Connect your leads database
Calendar: Auto-book appointments
WhatsApp: Trigger follow-up messages
Webhook: Notify your team of completed calls

Step 6: Test Your Agent

Use Bolna's built-in phone simulator to test your agent. Call it, have a conversation, and refine the responses until they sound natural.

Step 7: Go Live

Connect a phone number and your agent is live. It can now handle unlimited calls simultaneously — 24/7.

Real Example: Appointment Booking Agent

Here's a simple appointment booking agent I built for a dental clinic:

Agent Name: Dr. Smile AI

Goal: Book dental appointments

Language: Hinglish (Hindi + English mix)

Conversation Flow:

Greet the patient
Ask for their name
Ask what treatment they need
Check available slots
Confirm the appointment
Send a WhatsApp confirmation

The entire setup took under 2 hours. The clinic now handles 50+ appointment calls per day without any staff involvement.

Common Mistakes to Avoid

1. Vague System Prompts
Don't write: "Be helpful and answer questions"

Write: "You are [Name], a customer service agent for [Company]. Your only job is to [specific task]. Always [behavior]. Never [restriction]."

2. Ignoring Edge Cases
Test with difficult questions. What happens when a customer asks something off-topic? Train your agent to handle these gracefully.

3. Skipping Human Fallback
Always have a fallback to connect the caller to a human agent when the AI cannot help. This prevents frustrated customers.

4. Not Testing Enough
Test with 20-30 different scenarios before going live. Ask friends and colleagues to test it with unexpected questions.

Cost: How Much Does It Cost?

Platform	Starting Cost
Bolna AI	Free trial, then ~$0.05/minute
Vapi AI	Free trial, then ~$0.05/minute
n8n	Free self-hosted, $20/month cloud
Make.com	Free tier available, $9/month paid

For most small businesses, the total monthly cost for an AI agent is ₹2,000–₹8,000 — far less than a human employee.

Your Next Step

Start with a simple use case:

Appointment booking — easiest to build
FAQ answering — most common first project
Lead qualification — high ROI for sales teams

Pick one, build it this weekend, and see the results.

Omkar Vijay Ghate is an Agentic AI Architect based in Mumbai. He has built 16+ live AI agents for Indian businesses. Follow him on YouTube at @omilabs for more tutorials.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

The Future of Work: AI Agents Are Taking Over Repetitive Jobs (And That's Good)

Omkar Ghate — Thu, 29 Oct 2026 10:46:34 +0000

Originally published on Omkar Ghate's Portfolio

AI agents are transforming work as we know it. Here's an honest look at what's changing, what's not, and how to position yourself for the AI-augmented future.

The Shift That's Already Happening

In 2026, AI agents aren't a future prediction — they're a present reality. Companies are deploying AI agents to handle email triage, data entry, report generation, customer follow-ups, and scheduling. These aren't pilots. These are production systems.

What AI Agents Are Actually Replacing

The jobs being disrupted aren't "low-skill" jobs in the traditional sense. They're high-volume, repetitive cognitive tasks:

Data entry and validation
Basic customer service queries
First-pass document review
Appointment scheduling
Status updates and follow-ups
Report compilation

What AI Agents Can't Do (Yet)

Complex negotiation requiring human judgment
Creative problem-solving in novel situations
Building deep client relationships
Physical presence tasks
Ethical decision-making in ambiguous situations

How to Position Yourself

The workers thriving in 2026 are those who:

Understand AI tools — they know when and how to deploy agents
Specialize in oversight — they review and improve AI outputs
Focus on relationships — they do the human-facing work AI can't replicate
Learn prompt engineering — they know how to give AI effective instructions

The Indian Context

For India specifically, AI is both a threat and an opportunity. India's massive IT services sector is adapting rapidly. Companies that learn to deploy AI agents effectively will outcompete those that don't.

My Prediction

By 2028, every knowledge worker will manage at least one AI agent. The question isn't whether AI will change your job — it's whether you'll be the one directing the AI, or being replaced by someone who does.

Read the full analysis →

Written by Omkar Vijay Ghate, Agentic AI Architect based in Mumbai, India.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

AI Workflow Automation with n8n: The Complete Guide for 2026

Omkar Ghate — Thu, 29 Oct 2026 10:45:53 +0000

Originally published on Omkar Ghate's Portfolio

n8n has become the go-to tool for AI workflow automation. Here's everything you need to know to build powerful AI-powered automations in 2026.

Why n8n for AI Automation?

n8n gives you the visual workflow builder of Zapier with the power of code when you need it. Unlike other tools, n8n lets you self-host, connect to any AI model, and build complex multi-step agents.

Key AI Nodes in n8n

AI Agent node — build ReAct-style agents that can use tools
LLM Chain — chain multiple AI calls together
Vector Store nodes — connect to Pinecone, Qdrant, Weaviate
Memory nodes — give your agents conversation memory

Building Your First AI Workflow

Start with a trigger (webhook, schedule, email)
Add data enrichment (fetch context from your CRM or database)
Pass to AI Agent with tools
Route output to your destination (Slack, email, CRM update)

Real Business Use Cases

Lead qualification: Incoming lead → AI scores and qualifies → routes to the right sales rep
Customer support: Support ticket → AI drafts response → human reviews → sends
Content pipeline: Trending topic → AI research → draft → publish

Pro Tips

Always test with small data sets first. Use n8n's built-in error handling. Set up monitoring with webhooks to track workflow failures.

Getting Started

n8n is free to self-host. Try n8n.cloud for a managed version. Start with their template library and modify from there.

Read the full guide with screenshots →

Written by Omkar Vijay Ghate, Agentic AI Architect based in Mumbai, India.

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

AI vs Human Customer Service: The Real Comparison in 2026

Omkar Ghate — Thu, 29 Oct 2026 10:38:58 +0000

Originally published on Omkar Ghate's Portfolio

Everyone's debating AI vs human customer service. Here's the honest comparison based on real deployments.

Speed

AI wins. Zero hold time. Instant response. 24/7 availability. No shift changes, no breaks, no "let me transfer you."

Empathy

Humans win — for now. When a customer is genuinely distressed, human agents can read emotional nuance better. But the gap is closing fast.

Cost

AI wins massively. One AI agent handles what 50 humans handle, at 5-10% of the cost.

Consistency

AI wins. Human agents have good days and bad days. AI delivers the same quality every single call.

Complex Problem Solving

Humans win. Edge cases, unusual requests, multi-step negotiations — humans still handle these better.

The Real Answer: Hybrid

The best companies aren't choosing AI or human. They're using AI for:

First contact and triage
Simple FAQs and status checks
Outbound reminders and follow-ups

And humans for:

Complex complaints
High-value customer relationships
Escalations from AI

What This Means for Your Business

If you're running a customer service team in India, start with AI handling 60-70% of volume. Let your human agents focus on the 30% that actually needs them. Both sides win — and so does your customer.

Omkar Ghate is an Agentic AI Architect based in Mumbai. Read more

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

AI Customer Service India: Why Indian Businesses Are Replacing Call Centers with AI Voice Agents

Omkar Ghate — Thu, 29 Oct 2026 10:38:23 +0000

Originally published on Omkar Ghate's Portfolio

Indian businesses are quietly replacing traditional call centers with AI voice agents. Here's what's really happening on the ground.

The Call Center Problem in India

India has one of the world's largest call center industries — but it's expensive, inconsistent, and hard to scale. Attrition rates hover around 30-40%. Training costs are high. And customers? They're increasingly frustrated with long wait times and repeated explanations.

AI voice agents are changing this equation entirely.

What AI Customer Service Actually Looks Like

Modern AI voice agents can:

Handle thousands of simultaneous calls without hold times
Speak in Hindi, Marathi, Tamil, Telugu, and 20+ Indian languages
Integrate with your CRM in real-time
Escalate to humans only when needed

Real Results from Indian Companies

Companies like TVS Motors and Mahindra are already deploying AI voice agents for lead qualification, service reminders, and after-sales follow-ups. The results are striking: 60-70% cost reduction, 24/7 availability, and consistent messaging every single time.

The Objection: "Our customers won't accept AI"

This was true in 2020. It's not true in 2026. Customers care about speed and resolution — not whether it's a human or AI. If the AI solves their problem in 90 seconds, they're happy.

What to Implement First

Start with outbound use cases: payment reminders, appointment confirmations, service follow-ups. These are low-risk, high-ROI starting points.

Then move to inbound: FAQ handling, order status, basic troubleshooting.

Tools to Start With

Bolna.ai — best for Indian language voice agents
ElevenLabs — for high-quality voice cloning
n8n — for CRM integration automation

The companies that deploy AI customer service in 2026 will have a significant competitive advantage. The question isn't whether to do it — it's how fast.

Omkar Ghate is an Agentic AI Architect who has built 14+ AI voice agents for Indian enterprises. Read more on his portfolio

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

Can AI Be Creative? The Real Answer to AI vs Human Art in 2026

Omkar Ghate — Thu, 29 Oct 2026 09:32:59 +0000

Can AI Be Creative? The Real Answer to AI vs Human Art in 2026 — Omkar Vijay Ghate

Can AI Be Creative? The Real Answer to AI vs Human Art in 2026

By Omkar Vijay Ghate · 2026-08-25 · LinkedIn · All Articles

Last month, an AI-generated image won first place at an art competition. The creator, who submitted it in the digital art category, later disclosed that he had used Midjourney to generate it. The ensuing controversy split the art world: was this cheating, or simply using a new tool — the way photographers once had to defend photography against the accusation that it was "not real art"?

The question "can AI be creative?" is deceptively simple. It sounds like a yes-or-no question, but it is really several questions bundled together: Can AI produce surprising outputs? Can AI express meaning? Can AI have intent? Can AI experience the world? The answers differ dramatically, and untangling them reveals something important about both AI and human creativity.

What AI Can Do Creatively (And It Is Remarkable)

Let me start with what is genuinely impressive. AI systems in 2026 can:

Generate images of extraordinary technical quality — photorealistic, painterly, fantastical — from text descriptions. Midjourney, DALL-E 3, and Stable Diffusion XL produce outputs that routinely astonish professional designers.

Compose music in virtually any style, from Bach chorales to ambient electronic to Tamil classical raga, at a level that can be indistinguishable from human composition to non-experts.
Write poetry, stories, and scripts with genuine craft — controlling rhythm, metaphor, structure, and tone with sophistication.
Generate unexpected combinations that feel genuinely surprising — a visual or textual mashup that no human would have thought to attempt in quite that way.
Iterate rapidly — produce 50 variations of a concept in the time it takes a human to produce one, allowing exploration of a creative space that would otherwise be impossible.

These capabilities are real and consequential. They change what is possible for individuals and businesses in creative work.

The Philosophical Problem: What Is Creativity, Anyway?

The debate about AI creativity usually stalls because the two sides are using different definitions of "creative." Let me offer a framework that resolves this.

Type 1 Creativity: Combinatorial Novelty

This is the ability to combine existing elements in new ways — to see connections that were not obvious before, to juxtapose things that did not exist together. Most of what we call "creative work" in commercial and professional contexts is Type 1 creativity: a logo that combines two concepts in a fresh way, a song that merges two genres, a product design that adapts a material from one industry to another.

AI is genuinely excellent at Type 1 creativity. Diffusion models for images are, in a meaningful sense, combinatorial machines — they learn patterns across millions of existing works and generate novel combinations of those patterns. The results are often genuinely surprising and aesthetically compelling.

Type 2 Creativity: Meaning-Making

This is creativity that expresses something — a point of view, a felt experience, a response to the world, a personal truth. Frida Kahlo's self-portraits are creative not because they combine paint in novel ways, but because they express her experience of chronic pain, her Mexican identity, her feminism, her tumultuous relationship with Diego Rivera. The meaning is inseparable from the person who made the work and the life they lived.

AI has no Type 2 creativity. It has no experience, no point of view formed by living, no emotional truth to express, no stake in what it makes. When an AI produces a painting of "loneliness," it is generating visual patterns associated with the concept of loneliness in its training data — it is not expressing the experience of loneliness.

The deepest creativity is not about novelty of form. It is about authenticity of expression. AI can produce the form without the authenticity. Whether that is "creative" is ultimately a definitional question — but it is a meaningful distinction for anyone who cares about art as communication between human beings.

The Market Implications: What Changes for Creative Professionals

Whatever you believe about the philosophy, the market reality is shifting rapidly. For creative professionals and businesses, here is what is actually happening:

What AI Is Already Replacing

Certain categories of creative work are being significantly disrupted by AI, and pretending otherwise is unhelpful:

Stock photography: The market for generic stock photos is collapsing. Why pay for a stock image when you can generate exactly the image you need in 30 seconds?
Generic illustration: Simple icons, decorative illustrations, pattern design — AI can generate these faster and cheaper than human illustrators. The commodity tier of illustration work is AI territory now.
Basic copywriting: Product descriptions, social media captions, simple marketing emails — these are increasingly handled by AI with minimal human refinement.
Background music: Generic background music for videos, podcasts, and presentations is largely AI-generated now. The market for this category of music composition has fundamentally changed.

What AI Is Amplifying

For creative professionals who embrace AI tools, the results are transformative:

Speed: A graphic designer using Midjourney for concept generation can present 20 creative directions in the time it previously took to develop 2. The entire creative exploration phase is compressed.
Scope: A solo creative now has access to capabilities that previously required a team. One person can write, illustrate, compose music for, and animate a short film.
Iteration: The ability to rapidly test and refine means better final work. AI generates the options; the human selects and refines.

What Remains Distinctly Human

Creative direction: The vision, the brief, the judgment about which of 50 AI outputs is worth developing — this is human work and remains so.
Meaning and intention: Work that expresses a genuine human perspective, that communicates authentic experience, that takes a stand — this requires a human.
Cultural and contextual sensitivity: Understanding what a piece of work means to a specific community, in a specific moment, in a specific cultural context. AI lacks this grounded understanding.
Relationship-based creative work: An artist who knows a patron, a designer who deeply understands a client's brand history, a musician who performs live and responds to an audience — the relational dimension of creative work is irreplaceable.

The Indian Creative Market in 2026

India's creative industries — Bollywood, advertising, fashion, gaming, architecture — are engaging with AI tools at different rates and with different implications:

Advertising and Marketing

Indian advertising agencies are using AI extensively for production work — generating multiple visual variants for A/B testing, adapting campaigns across regional languages, creating product images. The creative strategy and conceptual work remains firmly human. The production pipeline is dramatically more efficient.

Film and Entertainment

AI is used extensively for VFX, de-aging, de-noising, and background generation in Indian film. The creative direction, performance, and narrative remain human. Bollywood has also begun using AI for music composition for background scoring and for generating initial story treatments — but final creative decisions are human-driven.

Gaming

The Indian gaming industry uses AI for asset generation extensively — terrain, textures, character variants, background art. Human artists design the core aesthetic; AI generates the volume. This has allowed Indian studios to produce higher-quality games at lower cost.

Architecture and Design

Indian architectural firms use generative AI for exploring spatial concepts, generating massing models, and testing façade variations. The structural engineering, client relationship, and final design judgment remain human.

A Practical Framework for Creative Professionals

If you are a creative professional navigating this landscape, here is a practical framework:

Identify where you are irreplaceable. Your perspective, your relationships, your cultural understanding, your judgment, your voice — these are yours. Lead from them.
Use AI for volume and exploration. Generate 50 concepts instead of 5. Test more directions. Explore more possibilities. Human time is now better spent on selection and refinement than generation.
Be transparent with clients and audiences. AI-assisted work is not lesser work, but it is different work. Honesty about process builds trust.
Keep developing your own creative eye. The ability to recognise quality, to articulate what makes something work, to direct AI effectively — these are skills that matter more, not less, as AI takes over generation.
Specialise in cultural depth. AI is broadly trained and lacks deep cultural specificity. Expertise in a specific cultural tradition, aesthetic language, or community makes you irreplaceable in ways AI cannot replicate.

The artists who will thrive are not those who refuse AI and not those who outsource their creative identity to it. They are the ones who use AI as a highly capable but aesthetically unintelligent collaborator — bringing the vision, the meaning, and the judgment while letting AI handle the generative labour.

The Deeper Question

Here is what I believe, after thinking about this carefully: AI creativity is a category of its own, and it is neither the threat nor the liberation that its most extreme advocates or critics claim.

AI produces outputs that can be beautiful, surprising, and useful. It does not produce work that matters in the way human creative work matters — rooted in lived experience, expressing genuine perspective, speaking from one human to another. The emotional resonance of great art comes from the sense that another consciousness made something, struggled with something, chose to show something of themselves.

That resonance is not replicable by a probability distribution. And it is not going away just because the tools are changing.

Want to Build an AI Agent for Your Business?

Omkar Vijay Ghate has built 16+ live AI agents across automotive, real estate, finance and retail in India. Let's talk about what's possible for you.

Get in Touch →

Frequently Asked Questions

Is AI-generated art considered 'real' art?

This depends on your definition of art. If art is about skill, novelty, and aesthetic appeal, AI-generated work qualifies. If art requires human intentionality and expressed experience, AI generates outputs that resemble art but lack what makes art meaningful. The more useful question for practitioners is: does this output serve the purpose? For many commercial purposes, AI-generated visuals serve the purpose excellently.

Will AI replace graphic designers and artists?

AI will replace the commodity tier of creative work — generic stock images, basic icons, simple layout work. It will not replace creative direction, conceptual development, brand strategy, cultural insight, or relationship-based creative services. The graphic designers most at risk are those who primarily execute technically rather than think creatively. The ones most positioned to thrive are those who use AI as a tool and lead with judgment and vision.

What Indian creative tools use AI effectively in 2026?

Adobe Firefly is the most professionally adopted AI image generation tool among Indian designers. Canva Magic Studio handles marketing assets. For video, Runway and Pika Labs are widely used. In music, Suno AI and Udio generate background music. For writing and copy, Claude and ChatGPT are the most common. Most Indian creative agencies have now integrated at least one AI tool into their production workflow.

How should I as an artist protect my work from being used to train AI?

You can use services like Glaze (which adds imperceptible noise to images that confuses AI training) and Nightshade (which adds adversarial data to training sets). The legal landscape is evolving — the EU AI Act requires disclosure of copyrighted training data, and several US lawsuits are working through the courts. In the meantime, watermarking, registering copyright, and publishing on platforms with clear AI-training opt-out policies (like ArtStation's AI opt-out) gives you some protection.

Can AI understand aesthetics and beauty?

AI systems can model human aesthetic judgments — they learn which images humans rated highly and generate outputs that tend to score well by those measures. This is meaningfully different from understanding beauty. AI has no subjective experience of beauty, no emotional response to art, no sense of what resonates. It approximates human aesthetic preferences from data. This approximation is often very good — good enough to be useful — but it is not understanding.

Originally published at omkar-portfolio-ydov.onrender.com

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai

The 36 Laws of Quiet Power: Book Review and How It Changed My Approach to AI Work

Omkar Ghate — Thu, 29 Oct 2026 09:32:57 +0000

The 36 Laws of Quiet Power: Book Review and How It Changed My Approach to AI Work — Omkar Vijay Ghate

The 36 Laws of Quiet Power: Book Review and How It Changed My Approach to AI Work

By Omkar Vijay Ghate · 2026-08-10 · LinkedIn · All Articles

I read a lot of books about strategy, persuasion, and power — partly because they are fascinating, and partly because the work I do (building AI systems for Indian businesses) is as much about human dynamics as it is about technology. You can build the most sophisticated AI agent in the world and watch it fail because the people who needed to adopt it felt excluded from the decision to deploy it.

The 36 Laws of Quiet Power arrived in my reading list through a recommendation from a mentor who prefaced it with: "This is not about manipulation. It is about understanding." I am glad I picked it up.

What the Book Is (And Is Not)

The 36 Laws of Quiet Power is not a self-help book in the conventional sense. It does not promise to make you more likeable, more charismatic, or more persuasive in the crude sense. It is more accurately described as a study of subtle authority — the kind that operates without loud assertion, that builds without confrontation, that endures without dependency on position or title.

The "quiet" in the title is its defining feature. The book's thesis is that the most durable forms of power are understated: the reputation built slowly through consistent delivery, the influence that comes from being genuinely useful, the authority that others grant rather than the authority you demand.

This immediately distinguishes it from books like Robert Greene's 48 Laws of Power, which tends toward a more cynical and manipulative frame. Quiet Power is interested in something subtler and, I think, more sustainable — the kind of influence that works over years and across relationships rather than in single tactical moments.

The Laws That Stayed With Me

I am not going to summarise all 36 laws (that would defeat the purpose of reading the book). But let me share the ones that genuinely changed how I think:

Law 7: Speak Last

The person who speaks first in any group sets the frame. The person who speaks last shapes the conclusion. This law is about the strategic value of restraint — waiting until you understand the terrain before committing to a position. In technical discussions, I used to jump in early with opinions. Now I ask more questions first. The quality of my conclusions has improved significantly.

Law 12: Scarcity of Presence

People value access to you roughly in inverse proportion to how available you make yourself. This is not about artificial unavailability or game-playing — it is about the genuine scarcity of deep attention. When I am with a client, I am fully with them. I do not schedule back-to-back calls. The experience of talking to someone who is genuinely present is now rare enough that it registers.

Law 19: Let Them Take Credit

This one is counterintuitive and difficult to follow. When a good idea you introduced gets adopted and the person who adopted it receives credit for implementing it, the book argues that this is actually the best outcome — because now the idea is theirs, which means they will defend and advance it. I have tested this in practice. An AI system that a client's team feels ownership over gets better support, more budget, and faster adoption than one that is seen as "the consultant's thing."

Law 23: The Power of the Unfinished

Leaving something open — a question unanswered, a thought incomplete, a space for the other person to contribute — creates engagement more effectively than delivering a complete package. In presentations, I now deliberately leave space. "Here is what we know. Here is what we are uncertain about. What do you see from your side?" This transforms a monologue into a collaboration.

Law 31: Know When to Withdraw

Strategic withdrawal is not defeat. Sometimes the most powerful move is to step back, let a situation evolve, and re-enter at a better moment. I have lost clients by pushing too hard at the wrong time. I have won them back by demonstrating patience and returning when the timing was right.

The Criticism Worth Voicing

I want to be honest about where I think the book is limited:

Some of the laws work better for people who already have baseline credibility — professional track record, network, cultural capital. For someone starting out without those advantages, "speak last" can result in being spoken over. "Scarcity of presence" can result in being forgotten. The book's wisdom assumes a level of recognised competence that, once established, allows the subtle plays to work. Without that foundation, the approach can backfire.

The book is also somewhat culturally located in a Western professional context. Indian workplace dynamics — relationships with seniority, the role of family connections, the difference between Mumbai and Chennai professional cultures — add dimensions the book does not explicitly address. Some laws require adaptation for the Indian context; a few need to be significantly modified.

And finally, some of the laws edge toward manipulation, even if the book frames them otherwise. "Let them take credit" can be generosity or can be calculated self-effacement for strategic gain. The difference matters, and the book is not always careful to maintain it.

How It Changed My AI Work

The most surprising application of this book has been in how I think about AI deployment in organisations.

Building an AI agent is 30% technical and 70% organisational. You have to understand who is threatened by the change, who stands to benefit, who controls adoption, and how to bring each stakeholder along without triggering defensive resistance. The laws in this book are, in effect, a guide to navigating that organisational landscape.

Law 7 (speak last) means I now do a lot more listening in initial client conversations before presenting what the AI can do. I understand the political terrain first.

Law 19 (let them take credit) means the AI systems I build are presented as the client team's systems, not mine. Their name is in the documentation. Their ideas shaped the design. They demo it to their leadership. I am the technical partner in the background.

Law 31 (know when to withdraw) has saved me from several bad projects. When the organisational readiness is not there, the most powerful move is to delay. A forced AI deployment into a resistant organisation creates lasting damage to AI credibility that takes years to repair.

The most important insight I took from this book: the technical quality of an AI system matters far less than the trust of the people who have to live with it. Building that trust requires exactly the kind of quiet, patient, relationship-sensitive work that the 36 Laws describe.

Who Should Read This

This book is particularly valuable for:

Consultants and advisors who need to influence without hierarchical authority

Technical professionals who are good at their craft but find organisational navigation difficult
People early in their careers who are learning how trust and credibility work
Anyone managing up — working with or influencing people above them in the hierarchy
AI and technology professionals deploying systems into human organisations

It is less useful for people who already have strong political instincts, or for very junior professionals who need to first establish basic visibility before playing the long game.

My Rating: 4/5

This is a thoughtful, nuanced book about a topic that most professional development writing treats superficially. It is not perfect — the cultural specificity is limited, the ethical line between influence and manipulation is sometimes blurry, and some laws assume a level of established credibility that beginners do not have.

But its core insight — that sustainable power is built quietly, through usefulness, patience, and genuine attention — is both true and undervalued in a culture that rewards loud self-promotion.

I have returned to several of these laws repeatedly in the two years since I read it. That is probably the best endorsement I can give a business book.

Frequently Asked Questions

What is The 36 Laws of Quiet Power about?

The 36 Laws of Quiet Power is a guide to subtle influence and understated authority — the kind of power that is built over time through genuine usefulness, patience, and attention rather than through loud assertion or manipulation. It argues that the most durable forms of influence are quiet ones, and lays out 36 principles for developing them.

How does The 36 Laws of Quiet Power differ from The 48 Laws of Power by Robert Greene?

The 48 Laws of Power is often cynical and tactical, presenting power as a game played for dominance. The 36 Laws of Quiet Power is more interested in sustainable influence rooted in trust and usefulness. It is less about winning individual exchanges and more about building a long-term reputation and the kind of authority others grant voluntarily rather than what you seize.

Is this book applicable to professionals in India?

Yes, with some adaptation. The core principles — patience, usefulness, restraint, letting others take credit, strategic withdrawal — translate well to Indian professional contexts. Some laws need adjustment for hierarchical workplace cultures where seniority matters more and direct disagreement with seniors is less acceptable. The book is worth reading critically and adapting rather than applying wholesale.

What is the most useful principle from the book for technical professionals?

Law 19 (Let Them Take Credit) is probably the most counter-intuitive and most valuable for technical professionals who often want their work recognised. For anyone deploying AI or technical systems into organisations, making the client team feel ownership over the system dramatically improves adoption and reduces resistance. The technical excellence fades into the background; the organisational relationship is what endures.

Should I read this book if I am early in my career?

Conditionally yes. The quiet power approach works well when you have established baseline credibility — a track record, a reputation, some recognised skill. Very early in a career, being too quiet can mean being overlooked. Read it with that caveat in mind: the principles become more powerful as your credibility grows. Start by reading the laws about building reputation and usefulness, then layer in the subtler ones as your career develops.

Originally published at omkar-portfolio-ydov.onrender.com

📚 Books by Omkar Ghate

The Human Pattern — A modern guide to understanding human behaviour, patterns and decision-making in the age of AI.

The 36 Laws of Quiet Power — A modern framework of influence, restraint and strategy.

Both available on Amazon India 👆

— Omkar Ghate | Agentic AI Architect | Mumbai