The honest version
Why I build.
Not the press release. The actual answer.
I got obsessed with startups early.
I didn’t have a grand plan. I just got deeply interested in how products get built — how someone goes from an idea to something real that people actually use. That pulled me toward startups, toward software, toward building things on the internet.
I started experimenting with code, APIs, and AI tools before I knew what I was doing. Most of it went nowhere. I built things nobody used. I launched products that failed within weeks. I spent weeks on systems that turned out to be completely wrong approaches.
Over the years, I’ve attempted around 20-25 startups — apps, tools, platforms, micro‑SaaS ideas. The vast majority never made it past the first few months. Some were bad ideas, some I executed poorly, and some were just ahead of their time for my skill level at that moment. But every single one left a lesson that I still use today.
At some point the experiments became real products. NovaX AI started as one of those experiments. So did Saathi AI. The difference between an experiment and a product is just that you kept going long enough to make it actually work.
The real reasons.
Why startups
A startup is the fastest way to learn whether an idea actually matters. You don’t need permission. You don’t need a title. You just build and find out. That directness is what I find compelling — feedback loops measured in days, not years.
Why AI
AI genuinely changes what’s possible to build as one person. Before, you needed entire teams to build what one person with the right LLM stack can now ship in weeks. That’s not hype — I’ve experienced it firsthand. That leverage is real.
Why I continue after failures
Because stopping doesn’t actually help. Every version that didn’t work taught me something specific about why it didn’t work. That knowledge compounds. A year of failed experiments is still a year of learning — it just feels worse in the moment.
Why execution over ideas
Ideas are everywhere. Everybody has ten ideas a week. The hard part is turning one into something real, then making it reliable, then making it useful to people who didn’t already know they needed it. That’s the actual game.
Divyansh Shukla
Divyansh Shukla is an entrepreneur, author, and AI builder focused on creating practical technology products for the next generation of users, creators, and businesses. With a strong interest in artificial intelligence, digital innovation, branding, and product strategy, he is building a reputation as a young founder working on accessible and future-focused AI solutions.
He completed his education at Delhi Public School, Lucknow, where his interest in technology, entrepreneurship, and problem-solving began to grow at an early stage. Alongside academics, he actively explored the rapidly evolving world of artificial intelligence, startups, and digital product building.
Divyansh is the founder of NovaX AI — an AI-focused platform built with the vision of making artificial intelligence more practical, accessible, and user-friendly. NovaX AI is positioned as a next-generation AI assistant platform designed to help users interact with advanced AI tools in a simple and efficient way.
Alongside NovaX AI, Divyansh is currently building Ghumakkad AI — an emerging AI-driven platform aimed at transforming how people explore, travel, and discover experiences. It combines AI with smart travel assistance, recommendations, and digital exploration tools to create a more personalized travel experience.
He is also an author who writes about innovation, AI, strategy, and the mindset required to build impactful products. His work is centered around the belief that innovation should be practical, scalable, and accessible to everyone.
Books authored.
Think Different. Build Different. Deploy Different.
The Strategic Playbook for Builders, Innovators, and Future Founders — a strategic guide for entrepreneurs and future founders focused on building meaningful products, developing unconventional thinking, and executing ideas effectively.
Written for those who want to move beyond conventional wisdom and build things that actually matter in the modern digital era.
AI for Everyone: The Practical Guide to Understanding Artificial Intelligence and Building Your Own AI Chatbot
A beginner-friendly guide designed to simplify artificial intelligence and help readers understand the foundations of AI systems and chatbot development in a practical way.
Covers how AI works, and walks readers through building their own AI chatbot — no prior technical experience required. For students, creators, and anyone curious about the AI revolution.
The failures.
Built multiple chatbot systems before NovaX AI. Most were too narrow in scope — solved one use case and nothing else. Abandoned them when I realized I was building features, not products.
First version of the NovaX backend was overengineered. Spent weeks building infrastructure for scale I didn’t have yet. Rebuilt it simpler. The simpler version was faster, more reliable, and easier to iterate.
Built things and assumed people would find them. They didn’t. Distribution is as much of a product as the product itself. Still figuring this one out.
Spent time on ideas because they seemed popular in the AI space, not because I actually believed in the problem. That never ends well. The only projects that moved forward were ones I genuinely cared about fixing.
Technical focus areas
AI & LLMs
- LLM integrations (OpenAI, Gemini)
- Prompt engineering & optimization
- Conversational AI systems
- RAG architectures
- Agent orchestration
- AI workflow automation
Backend
- Python
- FastAPI
- Cloudflare Workers
- REST API design
- Automation pipelines
- Database basics
Frontend & Web
- HTML, CSS, JavaScript
- Responsive design
- Performance optimization
- Technical SEO
- Web scraping / Selenium
- Cloudflare Pages
Infrastructure
- Cloudflare ecosystem
- Firebase
- Git & GitHub
- Deployment pipelines
- DNS & domain management
How it happened
Got obsessed with AI
Started experimenting seriously with LLM APIs and chatbot systems. Moved fast from tutorials to building real systems — most of which didn’t work at first.
Founded NovaX AI
Designed and deployed a general‑purpose AI platform independently. Built the full stack alone — backend, API routing, prompt architecture, and frontend. First real product.
Built Saathi AI
Created an AI health guidance system focused on accessibility. Different motivation than NovaX — built it because I genuinely thought it could help people who lack access to basic medical information.
10+ press features
Fox Story India, Borok Times, Entrepreneur Hunt, Hindustan Metro, UP18 News and more. Covered for a disciplined, execution‑focused approach to AI building from India.
Building Ghumakkad AI + scaling NovaX
Developing an AI travel assistant while growing NovaX AI’s capabilities. Still learning. Still shipping. Writing publicly about what I’m figuring out along the way.