cd /news/artificial-intelligence/my-interviews-were-terrible-then-ai-… · home topics artificial-intelligence article
[ARTICLE · art-116272] src=dev.to ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

My interviews Were Terrible. Then AI showed me How to Build

Sanjay R, a second-year computer science student at Rajalakshmi Engineering College in Chennai, overcame repeated interview failures by using Google's Gemini AI to learn full-stack development and build projects. He developed an early-stage application for identifying early indicators of autism, streamlined tasks with the Antigravity CLI, and began working on AXON OS, an AI-centric operating system concept. Sanjay now emphasizes making AI accessible and educating others about its risks and limitations.

read5 min views3 publishedAug 31, 2026

What if students all over India felt empowered enough to just build with AI?

Not because they memorized every complicated technical detail, but because they realized that AI could help them learn, test their theories, and turn an idea in their head into something tangible.

I'm Sanjay R, a second-year Computer Science student at Rajalakshmi Engineering College in Chennai.

For the longest time, I believed that AI was reserved for people with bigger resources, better mentors, and more experience than me. It took a series of failures and a mountain of self-doubt to finally shatter that belief.

My first year went off without a hitch.

I aced coding interviews, breezed through assignments, and generally felt like I was on the right track.

Then came year two. Things got tougher.

Club interviews and placement preparation committees came in wave after wave.

Each time, it ended the same way:

"Nice ideas, Sanjay, but you lack full-stack development skills. Come back when you know how to build end to end."

While I understood what they were saying, the repeated failures chipped away at my confidence. It felt like creativity and imagination were insufficient unless I could immediately translate every idea into a fully formed product.

I watched my peers soar through interviews and assumed they possessed an "X factor" I hadn't discovered.

I started pointing fingers at myself.

During the public launch phase of Gemini, I stumbled upon it.

I wasn't looking for a shortcut.

I was looking for an AI assistant that would allow me to ask "stupid" questions without fear of embarrassment.

The paradigm shift happened when I started pushing beyond basic prompts.

I delved into the why behind certain approaches.

Instead of simply asking for answers, I began asking:

Coupled with my familiarity with Google Colab and the Antigravity CLI, Gemini became an integral part of my learning journey.

Suddenly, my interview explanations were more coherent.

A sense of calm and then renewed confidence came to me.

I had found a tool that could meet me where I was and help me move forward.

Then, I began to build.

I developed an early-stage application aimed at exploring how AI could assist in identifying potential early indicators of autism.

I streamlined repetitive development tasks with the Antigravity CLI.

And I began working on AXON OS, an operating system concept where AI is at its core.

Then, something interesting happened.

Friends started asking me:

"How did you create this?"

That was my signal.

I not only learned how to build, but I also learned how to articulate and explain the process to others.

Don't get me wrong.

I don't know everything.

I get stuck frequently.

I am still learning and exploring.

But I remember very clearly what it feels like to have an idea and not know the first step.

I remember what it feels like to :

When someone asks me for help, I am not just giving them an answer

I first try to understand where they're stuck.

Then, I offer them a path forward.

It is the kind of guidance I desperately needed at the start.

My guiding principle is simple:

"You don't have to be a professional or a teacher to teach someone something. You only have to have the thought that something you know could serve someone else."

If I found a way to make a specific journey easier, I can certainly help illuminate the path for others. Making AI accessible is not as simple as giving someone a tool and saying:

"Go build."

It also means educating people about AI's inherent risks and limitations.

An application exploring early indicators of autism, for example, could cause significant distress if parents perceived its output as a medical diagnosis rather than an informational recommendation.

AXON OS, if poorly designed from a permissions perspective, could easily become a surveillance concern.

This is why i am passionate about discussing AI's:

These conversations shouldn't begin only after someone becomes an AI professional.

They should begin when someone first starts building.

We shouldn't require expensive courses or perfect English to start learning and exploring AI.

My goal is to create a learning environment accessible through free tools in both English and Tamil.

If I'm given the opportunity to be a Google Gen AI Ambassador, my objective would be to establish something with lasting impact beyond a singular event. I envision hosting monthly meetups in Chennai where students would gather to:

I also envision a vibrant Discord community where students can:

Partnering with colleges in Tamil Nadu, including my own institution, would be crucial in creating continued support and opportunities for students.

I've already begun replicating this within my own social circle.

Now, I want to scale up those grassroots efforts.

I believe an ambassador needs this:

A deep understanding of the students they are trying to inspire and guide.

As a second-year engineering student at a tier-two institution in Chennai, I didn't start with an expansive network or accomplished mentors.

I pieced things together through perseverance.

I faced rejection.

I dealt with self-doubt.

I struggled to turn ideas into working applications.

And because of that journey, I can deeply resonate with the student who thinks:

"AI is beyond my reach."

I want to be the guidepost that shows them otherwise.

Not by telling them that everything will be easy.

But by helping them take the first step.

The trajectory of India will inevitably be shaped by artificial intelligence.

The critical question is:

Some people may have a head start because of privileged access, resources, mentorship, or opportunity.

But no one should be left out of the conversation.

I've experienced failed interviews.

I've wrestled with self-doubt.

I've struggled to piece together simple applications.

The resources provided by Google, particularly Gemini, proved invaluable in helping me move forward.

Now, I want to pay it forward.

I want to be proof for other students that:

An idea can evolve into something tangible with the right tools, support, curiosity, and commitment to continuously learn and adapt.

Google aims to democratize AI.

And I aim to help bring that opportunity closer to students in my home state of Tamil Nadu.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @sanjay r 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/my-interviews-were-t…] indexed:0 read:5min 2026-08-31 ·