How I Got a Job Offer as a Fresher in an AI-Fearing Market — Without LeetCode or Club Leaderships on My Resume

I didn't have an impressive LeetCode profile.

I wasn't a club president.

I didn't have a stack of leadership positions on my resume.

If you've been anywhere near placement season this year, you know the vibe: an AI-scared job market, everyone grinding LeetCode, resumes stacked with club leaderships and internships. I had none of that. What I did have was a CPI that dipped before it climbed, one internship I was told to hide, and a platform I stumbled into almost by accident. This is that story.

The Dip Nobody Warns You About

First two semesters of B.Tech were easy — and that SPI doesn't even count toward your final CPI, so nobody sweats it. Sem 3 had just three subjects, still manageable. Then sem 4 hit: five subjects, a monthly exam cycle, two projects, and a hardware subject (8085 microprocessor) that I did not see coming. My CPI dropped from 8.52 to 8.13. My SPI that semester was 7.8 — the first time it had ever gone below 8.5.

To make it worse, this was before AI IDEs were really a thing. Most of the tools we casually use today either didn't exist yet or were still too new to be genuinely useful.

The Bounce Back

Sem 5 came with an even heavier load — five subjects plus three projects. This was the semester I decided I couldn't afford to slip again. So I started using every AI IDE I could get my hands on — Copilot, Cursor, Blackbox, Warp — and built three projects for my sem assignments across three different stacks: a Flutter app, a MERN project (which would later become ShelfTalk, revived and entered into GitHub's Finish-Up-A-Thon), and a .NET project.

Result: SPI went from 7.8 to 8.54.

8.52 → 7.80 → 8.54

CPI climbed back to 8.28. Not some massive academic comeback — but proof I could take a hit and recover.

Taking a Risk

Even with the recovery, I had no LeetCode profile worth mentioning, and I knew I'd be up against roughly 300 people for on-campus placements. So in sem 6, I took a bit of a gamble — I started applying for internships while still commuting to college every day.

I landed one: a paid, three-month, work-from-home internship at a small Bangalore startup. The first two months were fine. Then I became the only person in the IT department, got blamed for things I hadn't built, and watched the non-technical founder change UI flows every few days. The agency that had staffed a second developer alongside me cut that dev's hours down to four a day — which meant I was carrying 6+ hours on my own. My day looked like this:

6:30 AM — leave home for college
College — all day
5:30 PM — get back home
6 PM – midnight — work

No new hires. No real learning. Just "use Cursor" as the answer to everything.

I left five days before my sem 6 final exams. That's a story for another day — I'm not getting into all of it here.

My professor, the one who'd helped me through most of B.Tech, gave me one piece of advice afterward: don't put this internship anywhere. Not the resume, not LinkedIn. His reasoning — the moment it's visible, the interview stops being about your skills and turns into an interrogation about why the internship was incomplete.

Rebuilding, One Hackathon at a Time

With that internship off the table as a resume line, I knew I had to build something else to stand on. So I spent that summer vacation hacking — building out a stronger portfolio of domain-specific projects, drilling every system design concept I could, and deliberately switching tech stacks every time so nothing looked repetitive.

Somewhere in there, a hackathon I'd entered announced its winners just 45 minutes after the submission deadline closed. A lot of participants were furious and reached out to the organizers, who gave vague, spinning non-answers. Contacting the hosting platform didn't help either.

That frustration is what pushed me to actually explore an account I'd created on DEV.to back in January but never used. I already knew DEV had been acquired by MLH in February — and it clicked that this was the most authentic, professional space I could put my work in front of.

Finding DEV.to

I started small: the Gemma 4 Challenge first, then GitHub's Finish-Up-A-Thon, then the June Solstice Game Jam. By the time placements rolled around, I had a genuinely solid project portfolio behind me.

But portfolios don't get you shortlisted on their own. My resume kept getting filtered out — no club leadership, no internship listed. Standard rejection reasons.

Then, a few days before the Finish-Up-A-Thon results were even announced, DEV.to emailed me: I'd been made a Trusted Member, with moderation rights on posts.

This email arrived a few days before the results.

ShelfTalk — the MERN project from sem 5 that I revived for the Finish-Up-A-Thon — went on to place Top 10 of 500+ entries:

ShelfTalk — Books Don't Talk. We Do. (A College Project, Revived)

GitHub “Finish-Up-A-Thon” Challenge Submission

Dhruv Jani

Dhruv Jani

Dhruv Jani

Follow

Jun 7

ShelfTalk — Books Don't Talk. We Do. (A College Project, Revived)

#showdev #devchallenge #githubchallenge #webdev

9 reactions

9 comments

4 min read

So in a short window, I had a global challenge win, community recognition, and a trusted-member badge — three things that weren't there before. With just two changes to my resume, it started getting shortlisted. And once I cleared the DSA round for the first time, I had a strong hunch the rest — tech and HR interviews — wouldn't be as brutal. I was right.

The .NET internship, hidden as it was, still helped a little in HR rounds — my professor's rule was simple: only bring it up if they explicitly ask for past experience, and if they push further, tell them everything, just not on social media.

The Simform Day

Simform came to campus on a Saturday. I had a close relative's social event the very next day. Honestly, I walked in expecting to get knocked out in round 2 (DSA) — I was running on empty from back-to-back interviews all week and mostly just wanted to make it to my cousins in time.

I didn't get rejected. I made it all the way to the HR round, which stretched into the evening. I got the offer.

I ended up meeting my cousins a day late — but it was worth it when the results came in on Sunday and my TPO told me not to sit for any further placement drives.

No shit, Sherlock. I wasn't going to.

Simform was genuinely my best shot at this stage — after the internship I'd been through, what I wanted most was a company with an actual functioning HR department. That's rarer to find than it should be.

So was it LeetCode? No.

Was it club leadership? No.

It was proof — projects, a win, a badge — piling up until my resume finally said something worth pausing on. That's the story, start to finish: a CPI that dipped and recovered, one internship I chose not to talk about, a scammy hackathon that (indirectly) led me to the right platform, and a global challenge win and community recognition that changed what my resume could say. No LeetCode grind, no club title — just consistent building and the right places to show it.

If there's one line that sums up how I've come to see all of this, it's this shlok from the Bhagavad Gita:

कर्मण्येवाधिकारस्ते मा फलेषु कदाचन।
मा कर्मफलहेतुर्भूर्मा ते सङ्गोऽस्त्वकर्मणि॥

Translation: You have the right to perform your actions, but you are not entitled to the fruits of those actions. Never consider yourself the cause of the results, and never let attachment to results stop you from doing your work.

I couldn't control when a hackathon would turn out to be scammy, when a startup would fall apart, or when a random DEV email would land in my inbox right before results came out. All I could actually control was showing up and building the next thing. The outcome sorted itself out.

P.S. I still don't have a particularly impressive LeetCode profile. And honestly? I'm okay with that.