Hey there!
This will be the last blog for 2025, and I wish you a happy New Year in advance.
Sam Altman, the founder of ChatGPT and OpenAI, said this year that “the next revolution will be AI agents, AI agents building websites, writing content, posting and scheduling content and doing more.
I read this year’s 2025 YC funding statistics blog, which states that 359 companies among 420 were AI-based startups. Building AI agents is the next top skill to learn in software development.
This certainly opens up the way to grab your next job or to build your next idea.
I often asked a questions about how I learn new technology in programming, and I often say: YouTube(my favourite YouTuber for beginners) and online courses from any website, udemy for example, is my only source for a long time, and not to forget that I consistently read a lot of online articles/blogs/emails (like this one 😁).
AI agents are the future. We will all be using AI agents in the next few years, allow me to connect the dots
- 1990s: Neural networks emerge, enabling machines to learn patterns from data.
- 2000s: Language models evolve to predict the next word or token using neural networks.
- 2018–2021: Large Language Models (LLMs) appear, trained on massive datasets for deeper understanding.
- 2022: ChatGPT popularises conversational AI for everyday users.
- 2024: Agentic AI systems arise, capable of using tools and performing real tasks autonomously.
- 2025–2026: AI becomes embedded across platforms, automating workflows and reducing manual work.
Feel free to judge 😂
I often ask myself this question: What will the future website look like then?
One thing is certain: as a consumer and creator, we will be dealing with plain english and prompting a lot and explaining our ideas to one chat input. A few layers will be added, such as voice agents and image reading/perceiving LLM models that make a computer a Jarvis, which can talk, understand and perform actions.
One shot of imagination tells that probably each in future has something related to AI chat input, for examples
- SAAS websites will have an AI chatbot to answer queries, and agents to do manual work
- The documentation websites willhave AI chat answering doubts
- Health/Finance/CRM/ERP/E-commerce websites will have LLMs agents running in parallel to do tasks on users behalf
- Education websites will provide customised AI solutions for better learning, such as AI personalised quizzes, courses, deep research and scraping content
A few aggregators and social media platforms, such as ThereisAnAIForThat, Twitter, Reddit & Youtube helps me to understand and predict the future.
I often say this online is the future, be it at any business, online is the future, if one knows how to use online and social media, the money follows with a lot of opportunities. AI tools helping content creators are earning massive MRR not because they are the next trendy thing but because it actually works and it’s the most needed thing in future.
People connecting people, small communities to large, massive traffic websites like Twitter and Medium, all generating revenue because of people.
Boringcashcow is a simple aggregator website with almost 50k visitors per month, the real story of small online businesses generating massive revenue/profits. One can find tonnes of YouTube videos about how these boring small websites are earning millions, and one reason is that they solve a problem, at least one problem, for the people.
One of our clients for the React CRM template told me in a meeting that why don’t we should make a simple AI chat input that takes my prompts and makes changes in the existing template? And that’s why I add v0.dev and lovable.dev link for this React CRM template, helping our users to purchase and customise using the AI website builder.
An amazing way to help businesses is to provide them with a base to start, and then let the client/user make changes as needed.
Freedom is loved!!
Giving people freedom with tools always works for me as it build trusts, and that is why I always think gettemplate.website and other similar platforms are equally important in future
Gettemplate provides online websites for an inspirationa as well as a very first footstep to actually build an online business. Let me give a few more points
- 80+ FREE website templates, copy-paste the code or download the repository to further build your own
- 10+ production-level complete backend and database-integrated API,s such as CRM, Docs, SAAS builder templates to quickly start an online business
I’ve worked with 50+ clients in the past 5 years, and most of them have the same common requests: a few wants portfolio, a few want CRM, a few want ERP, a few want API automations, a few want forms, etc which gives an idea to launch gettemplate
Get your source code -> Build your idea, quickly and easily
This provides freedom and builds trust, which is loved by almost everyone.
Building an AI agent, how?
AI agents have a memory, a task orchestration and a monitoring agent as a system. One agent is a parent agent that decides the process for other agents, more like a manager, and other small agents are like Junior sales person or developers that perform tasks on behalf of the parent agent or manager agent.
People often miss this concept that AI agents are a system, not just a team of agents, a system that decides, performs and monitors the task.
For example, ChatGPT input is a simple one-agent that takes user questions and answers them, another agent is a web scraper that scrap internet for the latest data, such as weather reports, finance data, latest news, etc., and feeds that to the next AI agent to process the data and provide accurate answers.
I built the local version of Perplexity. I first created a web scraping API that scrapes any data from a URL, and another API for scraping internet URLs
- AI LLM to take a user question
- AI LLM use the web scraping API and web URLs scraping API to get the data from the internet under the hood
- Scraped data is fed into LLM to provide final answers; it’s a locally built Perplexity kind of application.
User adds a prompt -> AI agent understand prompt -> AI agent provide query to look on the internet => Web scraping API fetches the data using the query -> Feed the data into the AI LLM model along with intial use prompt => AI LLM finally provide answer to the user prompt
This is the entire way how Claude, Perplexity works under the hood, agentic behaviour is not much of a difference.
That would be enough for today
That’s it, see you in the next one
Shrey
Top comments (0)