← All writing

The year I started over

Nikunj Thakkar5 min read

#build in public#startups#lessons#entrepreneurship

Illustration: a fork in a road at sunrise with a small green seedling growing at the split

In January 2024 I did something I’d never done before: I told my whole network I was looking for a job.

I had left my full-time role feeling burned out and unsure of my next step. I’d been a founder, then a product leader at two B2B SaaS companies. On paper the next move was obvious: find another senior product role. So I posted #OpenToWork, listed what I’d done, and asked for help.

What actually followed was a year of freelancing, consulting, and small experiments. It ended somewhere I didn’t expect. This is the honest version of that year.

Asking in public

Posting #OpenToWork felt exposed. There’s a quiet assumption that experienced people don’t need to ask. But I’d decided that sharing the search publicly would reach people I’d never reach one message at a time.

The same week, I wrote about something that happened back in February 2016, the last time I’d left a job without a plan. I had spotted a post in a Facebook group from someone looking for a contract developer. I sent a DM. He asked for proof of work, and most of what I’d built was confidential. So I offered a guarantee instead: if he didn’t like the work, he wouldn’t pay.

He gave me a shot. They needed something ready for TechCrunch NY, and I delivered it in three days instead of the week they’d expected. Over the next six months that one DM turned into more than $40,000 of work. They eventually offered me a full-time role with cash and equity, but by then I had started DataOne.

Rereading that story in 2024 reminded me of something I’d forgotten while climbing a career ladder: the next opportunity is often one message away, if you’re willing to send it.

Saying yes to the work in front of me

The job search turned into consulting almost by accident.

In April I took on a freelance role as Head of GTM Strategy at Valuezen, working again with a former Whatfix colleague. Soon after, a friend who had just started a SaaS company asked how we were helping teams sell value. His team was struggling to close deals in a crowded market, so I asked to sit in on a sales demo.

The product was solid. The pitch was all features and pricing. We added real discovery questions to understand each customer’s problem, rebuilt the demo around the problems we found, and made a simple checklist to show where they were competitive. A few weeks later he told me they’d closed their first deal since the change.

Other work came the same way: product roadmaps for a US SaaS company, a careers portal for a Canadian university, go-to-market for a UK fintech entering India. None of it was planned. All of it taught me where I was genuinely useful.

Building small things again

The bigger change was that I started building again.

I’d been away from coding for a while, and AI had quietly made building fast. In late 2023 I’d started a content site on one of my spare domains. For two months, nothing. No visitors, nothing indexed. I kept publishing anyway. Then it tipped: more than 1,300 pages indexed, 23 organic clicks and 831 impressions in a single week. Tiny numbers, but they proved the compounding was real.

In July 2024 I built JobReferral over a weekend, a tool that lets companies post jobs with referral rewards and lets people earn by referring friends. Within days, 66 people had signed up as referrers and 51 recruiters had joined the waitlist. Then came the part nobody posts about: when you build solo, development is the easy bit. My days filled up with industry calls, outbound campaigns, Reddit threads, and waitlist emails. JobReferral stayed an experiment, and I didn’t pursue it further, but it taught me where the real work in a solo product sits.

Around the same time I restarted my podcast. After three episodes I had stalled, not on recording but on editing. Bringing on an intern to edit fixed it, and the podcast that became Startup OG got going again. The lesson was embarrassingly simple: the bottleneck is rarely the part you’re good at.

Then, over one weekend around the new year, I built FormSubmit, a small form backend for landing pages, because I was tired of building the same backend for every client site. A few weeks later I got my first payment from a complete stranger. There’s a particular joy in that, a tool you built for yourself turning out to help someone you’ve never met. Around then I also got buyout interest in a different small tool I’d built and never even launched.

Finding the thread

By the end of 2024 I could see the pattern. I loved helping founders take an idea from nothing to something real. I was good at the messy middle where product, technology, and go-to-market meet. And I liked doing it with people who were serious about building.

So on January 1, 2025, we launched UpSurge Ventures, a product studio for founders who want to turn ideas into real products. The first week: five calls booked, two warm leads, and a second proposal sent. On day seven we closed our first deal, partly because we met the customer face to face. By January 20 we’d closed two projects. A few weeks later Upahar Sood joined, and FormSubmit kept quietly picking up paid users.

A year after leaving my job burned out, I was building again.

What that year taught me

  • Asking for help in public works better than it feels. The discomfort lasts a day. The reach lasts much longer.
  • Say yes to adjacent work. Consulting showed me what I was actually good at faster than any amount of thinking about it.
  • Small experiments compound. A content site, a referral tool, a form backend. None of them were the plan. Each one taught me something the next one needed.
  • Burnout is information, not a verdict. I needed a year that didn’t look like a career move to find the work I wanted to do next.

If you’re in your own crossroads year right now, I’m happy to talk: me@nikunjthakkar.com.