Every Wednesday at Dezerv, we hold a product and design open house. Members of our product, design, and growth teams showcase new features and design directions to anyone interested. It’s optional but always valuable – people walk away having learned something new.
While I love attending these sessions, travel and client meetings often get in the way. Last week, I finally made it, and I’m glad I did.
One of our senior product leaders presented how we’re integrating AI across Dezerv. The highlight was an AI Voice Agent trained on Dezerv’s knowledge base. It functioned almost exactly like a relationship manager. What surprised me most wasn’t just that it worked – it was indistinguishable from a human. The speed of conversation, accent, accuracy, contextual understanding, and natural sentence construction were remarkable.
Even more astonishing? This colleague, who has zero coding experience, built this in just four hours.
This experience sent me down a rabbit hole exploring how AI-powered software will transform business building. We’re entering an era where non-technical professionals can deploy sophisticated AI agents to solve real business problems.
Welcome to this week’s Create Wealth blog. I’ll be covering AI agents and how they can leverage ChatGPT and other LLMs in ways that you and I never could, on our own. I believe it will fundamentally transform how India’s wealth creators work within the next 12-24 months.
What are AI Agents?
In very simple words, an AI Agent is software that uses AI to accomplish a goal that requires multiple steps.
Think of AI Agents as digital assistants that go beyond simple commands. These systems can work independently to handle complex tasks that normally require multiple steps and human judgment.
It has three main pillars:
- It must use an AI Model (ChatGPT, Claude, Llama, Gemini, etc)
- It must know how to use a tool (like a web browser, website, Google doc, etc)
- It must have memory, meaning it should be able to remember the original task across multiple tools and platforms it is using over an extended period.
Some AI Agents are fully autonomous—they can operate for extended periods without supervision, accessing various digital tools to complete assignments just as a human assistant would.
Others follow more structured workflows, performing specific sequences of actions based on predefined rules.
The pace of AI growth makes Moore’s Law look slow. Gordon Moore noticed in 1965 that computer power doubled every 18 months as we packed in more transistors. This pattern held steady for decades. But AI is evolving much faster – doubling in capability every 6 months rather than every 18.
For businesses, this means you can’t wait to adopt. Even a one-year delay could leave you far behind competitors who are riding this rapid wave of advancement.
Evolution of AI Agents
Past: AI agents began as simple rule-based systems like Microsoft’s Clippy (1997) and early chatbots. They followed predetermined scripts and couldn’t handle unexpected inputs. IBM’s Watson winning Jeopardy (2011) showed the potential for systems that could use information to answer questions, albeit in limited ways.
Present: Today’s AI agents like GitHub Copilot and ChatGPT plugins can use tools to complete specific tasks. They can make restaurant reservations, short-list hotels based on your preferences, send reservation emails based on your itinerary, and even speak to sales leads as first calls to reduce business development costs. Companies are using them for research, customer service, and to automate repetitive tasks.
Here’s another real example, GigaML – An AI Agent workflow company is helping Zepto address one million customer support tickets every month! That is insane. It has saved Zepto from hiring 500 customer support people. Imagine the cost savings.
Future: Soon, AI agents will continuously monitor your schedule like Jarvis from Iron Man, responding to emails that require simple observations. Quick commerce platforms like Instamart and Blinkit will likely build agents that integrate with your weekly meal prep menus and automatically order groceries. The technology will become more reliable, but human oversight will remain essential for critical decisions.
Teams will become hybrid – not in terms of location but composition, with humans and AI agents working side by side. Perhaps the most revolutionary will be agents using other AI agents, creating complex systems where specialized agents collaborate on tasks too intricate for a single agent to handle.
But do you know which company has single-handedly driven this AI boom?
Nvidia.
“When everyone digs for gold, sell shovels,” perfectly encapsulates Nvidia’s strategy to reach their current USD 3.3 trillion market capitalisation. Since their GPUs are essential for AI computations, the company supplies them to top firms like Google, Microsoft, Amazon, and Oracle, generating billions in revenue.
Once the current AI agents can consistently execute tasks, they will begin with their capacity to innovate, generate and explore new intellectual directions, just as humans approach problem-solving and creativity.
AI agents are gaining traction quickly across an array of business applications—and the market for AI agents is expected to grow at a 45% CAGR over the next five years.
By 2030, AI agents are expected to –
- handle over 80% of customer interactions.
- automate up to 70% of office work.
- make 15% of decisions autonomously.
How do AI Agents work?
AI agents go beyond automation.
Unlike basic automation tools that perform the same action repeatedly, AI Agents can adapt to changing circumstances, make decisions when faced with new information, and coordinate multiple systems to achieve their goals.
Anthropic, the company behind the popular LLM Claude, classifies these systems as “agentic systems” while making a key architectural distinction:
- Workflows are like following a recipe. The AI follows fixed steps that someone created in advance.
- Agents are like having a personal chef. The AI decides for itself what steps to take and which tools to use to get the job done.
Both can help you complete tasks, but agents have more freedom to figure things out on their own.
AI agents observe their environment, leverage large language models for planning, and access connected systems to take action and accomplish goals.
Observe: AI agents constantly collect and process information from their environment including user interactions, key performance metrics, or sensor data. They can retain memory across conversations, which provides ongoing context across multi-step plans and operations.
Plan: Using language models, AI agents autonomously evaluate and prioritize actions based on their understanding of the problem to be addressed, goals to be accomplished, context, and memory.
Act: AI agents leverage interfaces with enterprise systems, tools, and data sources to perform tasks. Tasks are governed by the plan delivered by a large language model or small language model. To execute tasks, the AI agent may access enterprise services (such as HR systems, order management systems, or CRMs), delegate actions to other AI agents, or ask the user for clarification. These intelligent software agents have the ability to detect errors, fix them, and learn through multi-step plans and internal checks.
How AI Agents will help you in your professional and personal life
Professional example: Client meeting preparation for a business consultant
A busy consultant needed to prepare for back-to-back client meetings across different industries. A process that once consumed her evenings is now handled by an AI agent, delivering personalized briefings an hour before each meeting. Here’s how it works:
1. AI agent gathers client updates: The agent scans recent news, LinkedIn activity, and company announcements related to the upcoming client.
2. AI agent reviews past interactions: The agent summarizes previous meeting notes, outstanding action items, and email exchanges from the last 90 days..
3. AI agent prepares talking points: The agent drafts 3-5 discussion topics based on industry trends relevant to the client’s current challenges.
4. AI agent delivers mobile-friendly brief: 60 minutes before each meeting, the agent sends a concise briefing to the consultant’s phone, allowing for quick review even between meetings.
Personal example: Family travel planning for a time-pressed executive
A CEO with limited family time wanted to plan a meaningful vacation without spending hours researching destinations. A process that typically consumes weekends now takes 30 minutes with an AI agent handling the details. Here’s how it works:
1. AI agent suggests destinations: Based on school calendars, weather patterns, and family preferences, the agent recommends 3 suitable destinations.
2. AI agent builds itinerary options: Once a destination is selected, the agent creates daily plans balancing activities for different family members’ interests.
3. AI agent handles bookings: After approval, the agent secures reservations for flights, accommodations, and key activities requiring advance booking.
4. AI agent creates a family-ready guide: The agent prepares a shared document with confirmation numbers, contact information, and day-by-day plans that the entire family can access.
Will AI kill SaaS?
The venture capital world is experiencing a seismic shift. For the past two decades, over 40% of VC money flowed into SaaS companies. That trend is changing.
Y Combinator predicts more than 300 unicorns (companies valued at USD 1+ billion) will emerge from the AI Agent space. This isn’t just speculation—it’s already happening.
Midjourney, an AI image generator that many artists and designers view with concern, generates USD 200 million in annual revenue with just 40 employees and no external investor funding.
The economics are staggering. AI Agents break the traditional correlation between revenue and headcount that has defined the SaaS business model. Midjourney’s USD 5 million revenue per employee dwarfs even the most efficient traditional software companies, which typically generate USD 300,000 – 500,000 per employee. Their service, accessible through Discord with packages ranging from USD 10 to USD 120 per month, demonstrates how AI companies can scale revenue without scaling the workforce.
We’re entering an era where AI Agent companies will systematically replace vertical SaaS solutions for mundane but essential functions: MIS reporting, data entry, annotation, quality control, and customer support. These tasks have always been necessary but represent significant cost centers for businesses.
Perhaps most revolutionary: AI Agents have the potential to create the first one-person billion-dollar company. I believe we’ll see this happen within the next decade. The combination of powerful AI models, minimal infrastructure costs, and global reach makes this not just possible but inevitable.
Are AI agents looking to take our jobs?
While people are worried that AI agents might take their jobs in the future, I believe this has already started. While scrolling Reddit, I came across this very interesting job description, specifically looking to hire an AI agent instead of an AI engineer.
Want to build your own AI Agent?
If you’re curious about creating your own AI agent, I’ve found an excellent resource to get you started. David, an AI enthusiast and creator, has put together a straightforward 40-minute tutorial that walks through building an AI agent using n8n, a no-code tool.
This tutorial is perfect for beginners and requires no programming experience.
I encourage you to watch it just to understand how these systems work under the hood. Maybe even use a weekend to build your first AI agent.
If you are like me and want to explore the full potential of AI agents before creating your own agents specific to your tasks and industry, let me introduce you to a library of pre-built agents, select one that fits their needs, and deploy it instantly.
Did you know about Agent.AI?
If building your own AI agent seems like a big, one great place to start is agent.ai, created by Dharmesh Shah.
I’ve long admired Dharmesh, co-founder of HubSpot—one of the world’s most successful SaaS companies.
Dharmesh, an early investor in OpenAI and self-proclaimed “biggest fan” of the technology, has launched agent.ai as “The #1 Professional Network For AI Agents” (and humorously notes it’s “also, the only professional network for AI agents”).
It’s essentially LinkedIn for AI agents—a place where specialized digital workers can be found and put to work.
Some of my favourite AI agents from agent.ai include:
1. Company Research Agent – I just have to paste the URL and the agent scans the entire Internet to create a company profile that would have taken an analyst several hours.
2. Ideal Customer Profile Builder – Perfect for founders and marketers who want help drafting an ICP for their products without extensive market research.
3. Video Script Generator – For crafting engaging script copy for your next video without spending on hiring a copywriter.
This move by one of SaaS’s most successful entrepreneurs speaks volumes about where the industry is heading.
How to leverage this opportunity?
Immediate opportunities
Even if you’re just starting, there are concrete steps you can take now:
- Experiment with available tools. Sign up for trial accounts with major platforms and test their capabilities against your specific needs.
- Identify quick wins. Look for simple, contained processes that could benefit from agent assistance without major integration challenges.
- Build internal knowledge. Assign team members to stay current on agent capabilities and case studies relevant to your industry.
- Engage vendors early. Even before you’re ready to buy, engage with leading vendors to understand their roadmaps and implementation approaches.
Long-term strategic considerations
As you move forward, keep these strategic considerations in mind:
- Workforce planning needs to account for the changing role of human workers as agents handle more routine tasks. Focus on developing uniquely human capabilities in your team.
- Process redesign may be necessary to fully capture agent benefits. Don’t simply automate existing processes; rethink them for the agent era.
- Data strategy becomes increasingly important as agents rely on high-quality, accessible data. Invest in data governance and integration to maximize agent effectiveness.
- Competitive monitoring is essential as agent adoption accelerates. Track how competitors use this technology to avoid falling behind.
In summary
AI agents represent the natural evolution of business automation. They combine the judgment capabilities of AI with the ability to take action across systems. For business leaders, they offer a once-in-a-generation opportunity to reimagine how work gets done.
The companies that thrive in the coming decade won’t be those with the most employees, but those that most effectively combine human and artificial intelligence. The time to start building that capability is now.
While the technology will continue to evolve, the fundamental shift is already clear: AI is moving from a tool we use to an assistant that works alongside us. For India’s wealth creators, this shift offers both an operational advantage today and an investment thesis for tomorrow.
Join over 300,000 wealth creators who receive our weekly finance, business, and wealth creation insights. Subscribe to the Create Wealth newsletter here – https://bit.ly/dezerv_newsletter
Disclaimer: The material is for informational purposes only and should not be interpreted as soliciting, advertising, or providing any advice. This material is based upon information sourced from reliable third-party sources, however, Dezerv does not represent that it is accurate or complete and it should not be relied upon as such. All trademarks, brand names, projects, and service names used herein are for identification purposes only and are the property of their respective owners. Use of these names and trademarks does not imply endorsement.