Building an AI Product in 70 Hours: A Story of Belief, Struggle, and Starting Without Code
- Arunava Chakravarty
- Apr 1
- 3 min read

I didn’t start with code.I started with a question.
“What if I could build something… even if I don’t know how to code?”
For years, I had been reading about AI—how it’s changing industries, replacing jobs, redefining everything.I would scroll, read, agree… and then move on.
Until one day, something felt uncomfortable. I realized I had become a consumer of ideas, not a creator.
And that day, I made a decision—“Enough reading. Let’s build something.”
The Idea: Where It All Began
I’ve spent decades working in program governance, delivery, and large-scale execution.And one problem kept repeating itself everywhere:
The gap between ideas and execution.
People had clarity.People had experience.People knew what needed to be done.
But they couldn’t build. Because they didn’t know how to code. That’s when the question hit me again:
“If coding wasn’t a barrier… what would I build?”
The answer came instantly: Recruitment.
I had seen HR teams struggle—hours spent reading resumes, subjective decisions, inconsistent evaluations. I knew what a better system looked like. I just didn’t know how to build it.
The Fear: The Hardest Step
Let me be honest. The hardest part was not technical . It was this moment:
Standing at the edge of something unfamiliar… and choosing to start anyway.
I am not a developer.I have never written production-level code.
And yet, I decided:
“I will build this.”
The Journey: Where Things Got Real
What followed was not smooth. Not even close.
Tools broke
Integrations failed
Outputs didn’t make sense
Things worked one moment and collapsed the next
There were many points where quitting would have been easier. And this is where most people stop.
But something shifted in my mindset.
Instead of asking:
“Why is this not working?”I started asking: “What is this teaching me?”
Every failure became a step forward. Not because it worked—but because I didn’t stop.
The Turning Point: Understanding AI Differently
At some point, I realized something powerful:
AI is not here to replace thinking but It is here to remove barriers.
I didn’t need to become a developer. I needed to become:
Clear in thinking
Structured in approach
Patient in iteration
AI handled:
Code generation
Debugging
System design
Output formatting
But it never decided what to build. That was still me.
The Build: What Came Out of It
What I Built
The AI Recruiter Dashboard is a live, deployed, web-based platform that does the following:
▸ JD Validation : Validates job descriptions against liveIndian industry benchmarks — Naukri database — ensuring JDs reflect what the market actually expects.
▸ Resume Scoring : Scores resumes across technical skills, certifications, and soft skills against validated requirements, with category-level breakdown.




▸ Behavioural NLP : Analyses candidate interview responses using natural language processing — scoring confidence, collaboration, problem-solving, and growth mindset from written answers.


▸ Salary Benchmarking : Compares candidate salary expectations against Indian market data by role and experience level.

▸ Professional Reports : Generates downloadable reports in Word, PDF, and Excel — ready for the hiring team with charts, scores, and hiring recommendations.
Access Management : Full recruiter registration, admin approval, and access code delivery via email — with a 3-use free tier and upgrade path.
Roadmap : For the candidate if HR wants to evaluate their internal candidate and plan for their future growth
The Realization: What This Journey Actually Meant
This was never just about building an app. It was about breaking a belief.
The belief that:
“You need technical skills to build something meaningful.”
What I learned instead:
You need clarity
You need persistence
You need courage to start without knowing everything
The Mindset Shift: The Real Outcome
This journey changed how I see things:
1. Experience is still your biggest strength
AI cannot replace real-world understanding.
2. Barriers are not gone—but they are lower
Coding is no longer the gatekeeper.
3. Iteration beats perfection
Progress > perfection.
4. Belief is the starting point
If you don’t believe you can build—you won’t even try.
5. AI rewards action
Not knowledge. Not theory but Action.
The Bigger Message
If someone like me—with zero coding background—can build a working AI product…
Then the question is no longer:
“Can I build?”
The real question is:
“What am I still waiting for?”
What’s Next
This is just the beginning. Because now I know:
Ideas don’t need permission anymore.
Execution doesn’t need coding anymore.
And AI is not a tool… it’s a mindset shift.
Explore the Application
Live Demo:
Disclaimer:
This is a proof-of-concept application built using free-tier AI tools to demonstrate a no-code development approach. It is not intended as a production system. Due to the limitations of free-tier infrastructure, users may experience slower performance and occasional inconsistencies.



Comments