After riding the rollercoaster of three startups, I find myself in a space of reflection, trying to untangle the threads of what I’ve learned and how I’ve changed. Looking back, it’s hard not to feel the weight of disappointment — most of my ventures didn’t turn out the way I had imagined. Some failed outright. Others ended in small victories that never felt big enough to celebrate. And yet, even in the midst of those frustrations, there’s pride. Pride not for the money made or the milestones hit, but for the sheer fact that I dared to dive in, to risk, to build, and to persist. These experiences have shaped me more than any book, course, or advice ever could. They’ve taught me patience, resilience, humility, and, perhaps most importantly, self-awareness.
It’s strange to try to describe the feeling — it’s messy, layered, and almost impossible to name. There’s a sense of loss for the things I thought I’d achieve, frustration at the mistakes I made, and exhaustion from the relentless grind. But alongside that is a deep, quiet satisfaction, a recognition that I’ve been through something few people willingly endure. Each misstep, each late night spent debugging or strategizing, each tough conversation with co-founders or customers, has left an imprint on me. It’s like walking through fire: it burns, it leaves scars, but it also tempers you.
And maybe that’s the essence of this journey — not the outcomes, not the numbers, not even the fame or recognition, but the process itself. The lessons learned, the ways I’ve grown, the clarity gained about who I am and what matters to me. It’s a journey I wouldn’t trade, even for a guaranteed success, because the failures taught me more than any victory ever could.
This is just the beginning of that reflection, a way to map the terrain of my experiences and the patterns that emerge when I step back. And as I sit here thinking about where I’ve been and where I might go next, I realize that the ride isn’t truly over — it’s only paused, giving me space to process, to rest, and to prepare for the next challenge.
Before diving into startups proper, I want to reflect on my SaaS projects. Honestly, they’ve been some of the most formative experiences in my journey. They weren’t huge successes in the headlines sense, but each taught me something about building, experimenting, and running a product.
The first project was unofficial — I was selling a wrapper around a wrapper for the WhatsApp API. Essentially, I provided access to the API without users having to go through all the Meta verification steps themselves. I charged around 12 a month, which in hindsight was too high; most users’ real usage only cost me 2–$3. At the peak, I had about 20 users, but many weren’t very active. Eventually, the user base dropped to just two paying users. I didn’t want to take advantage of them, so I decided to close it.
This project taught me a lot about pricing, user behavior, and responsibility. I didn’t do any marketing or SEO — I didn’t even know how — and I was awkward about promoting it, which I still am to this day. But the experience of building, maintaining, and eventually responsibly shutting down a product gave me lessons I couldn’t have learned from a textbook: about ethics, customer trust, and knowing when to step back.
Technically, it wasn’t hard. Backend logic, checking payments, and giving access — pretty straightforward. I realized early that a lot of SaaS projects are more about understanding users and logistics than the coding itself. Anyone could technically build it; the challenge was in sustaining and growing it.
The next project, Nexus, has honestly been one of my favorite creations. Unironically, I built it for GitHub, though I never actually put it on GitHub itself — and somehow, I lost the source code. I still have this itch to rebuild it, because I’ve never liked how GitHub displays analytics for codebases.
The problem started with a simple limitation: I didn’t pay for GitHub Pro, so I couldn’t access full repo analytics for my own code. I wanted to see meaningful stats — lines of code, contributions, patterns — so I decided to build my own system. It started small, just counting lines of code, but I kept improving it. Eventually, I thought, why not turn it into a SaaS? I added Stripe for payments, built the necessary authentication flows, and handled user permissions carefully so I could analyze their repos ethically.
Technically, it was a great challenge. I had to get GitHub users authenticated, pull data correctly, handle analytics, and build visually appealing cards and summaries. I looped through features again and again, refining it until it looked and felt like one of the best projects I had ever made. I did get some users, but not many — again, marketing wasn’t my strong suit. Still, this project taught me a ton about product design, data analytics, and building something that both works and feels polished. Even now, I think about rebuilding it one day, because it was a product I genuinely admired.
Then there was one project that’s honestly the most embarrassing: I got zero users. I priced it around 3–4 per month, and it was only meant for Nepal, so the market was tiny. The idea was a price-tracking tool for online shopping — you could track unlimited products, get email alerts when prices dropped, and even see which website had the cheapest version of the item. It was actually quite functional. I stored all the data, ran it on a cheap virtual server, and even built a web extension. The project didn’t cost me more than a dollar to run, and I learned so much in the process: web scraping, rotating IPs, handling multiple sources, and learning Next.js on the fly. Looking back, even if nobody used it, it was an incredible learning experience — a hands-on crash course in building web apps and understanding real-world data pipelines.
Then came MetaFeedbackEngine — my most recent SaaS project, around a year or so ago. Unlike the others, it had a $20/month pro plan and no free plan. The idea was to build a system that critiqued your own decisions, codes, journals, and tracked whether your goals were met. It was ambitious: I had to build an Electron app, integrate with web systems, and run costly models long before cheap AI tools existed. Back then, the running costs were brutal — I often lost money keeping it alive. It was the hardest project technically, but also the one where I learned the most about scaling complexity, running costs, and user friction.
The earlier SaaS projects were easier and somewhat profitable, but MetaFeedbackEngine was a real test — both technically and financially. Each of these projects taught me something unique: pricing vs. usage, technical problem-solving, handling cost vs. revenue, and the harsh reality that building a good product doesn’t automatically bring users.
There are more projects I’ve worked on, and more lessons from them, but that’s a story for another time. For now, these SaaS projects represent my first real foray into creating products that, even if not wildly successful, were deeply educational. They gave me confidence, technical chops, and a glimpse of the kind of work I want to pursue in the future.
Startup
One of my earliest startup experiences began when a friend messaged me: “Let’s build a startup. I have an idea.” He invited me to join as a co-founder, and I jumped in. The plan was ambitious — we wanted to build a suite of developer tools. One example was a JSON generator: developers could configure responses on a dashboard, and whenever they hit that endpoint, it would return exactly the mock data they needed. That was just one of the 100+ tools we eventually built.
By the time we were halfway through, I had built around 50 tools myself, while my co-founder had created 70–80 on his own. Maintaining the entire suite became overwhelming. The codebase was massive, one of the largest I’ve ever worked on, and each tool came with its own quirks, dependencies, and edge cases. Fixing bugs was hard, especially as the scale of the app grew — abstraction and proper organization would have helped, but we didn’t implement it at the time. The size of the codebase alone made every change a challenge.
Financially, the startup was even tougher. At our peak, we had roughly 356–370 paying users per month, but the revenue barely covered costs. Hosting and scaling expenses were huge, and we were also dealing with daily DDoS attacks. For three months, we had to put in our own money just to keep the project running. We even tried raising prices — some users accepted, some didn’t — but the losses continued. The business just wasn’t sustainable.
All marketing was handled by my co-founder, who built publicly and shared updates, bringing in users organically. I didn’t need to touch marketing at all, which allowed me to focus entirely on building.
Looking back, the project was a whirlwind of learning. I learned how to architect a large-scale codebase, manage multiple interconnected tools, and deal with the challenges of technical debt. I also learned a hard truth: even impressive technical output isn’t enough to sustain a startup. Scale, infrastructure, financial planning, and business strategy are equally critical. The experience was exhausting, frustrating, and at times overwhelming — but deeply educational.
The second startup was, honestly, my weirdest idea at first. The initial concept was… questionable: I thought we could build a platform where users could sell their own data to us, and we’d resell it to other companies. Looking back, it was a bad idea — and it sparked tension within the team almost immediately.
I reached out to the same friend from my first startup and invited another friend to join. Around that time, GPT-4 launched, and the hype was everywhere. It wasn’t quite the GPT-3 moment, but it got us thinking: maybe we could pivot. After some discussion, we abandoned the data-selling idea and decided to build an AI-based solution.
Our first pivot aimed at fitness: an AI agent that would analyze all the data a user provided and generate personalized fitness recommendations. It was ambitious, but none of us were confident in our approach, and the results didn’t meet our expectations. We tried again, pivoting to a broader idea — an AI system that could analyze personal data and give recommendations across different aspects of life.
Unfortunately, this pivot also failed to produce the outcome I had hoped for. Despite the energy and effort we poured into the project, it never gained traction, and eventually, we had to close it. The experience was incredibly frustrating. I was burned out from the pivots, and it highlighted just how difficult it is to find the right idea, even with promising technology and a motivated team.
What I learned from this experience is that good timing, clear focus, and a realistic understanding of what can actually be delivered matter as much as the technology itself. Even with cutting-edge AI, a product can fail if the problem isn’t clearly defined or if the execution is scattered.
The third startup — the one that truly tested me — was, without a doubt, the most memorable. It was wild, intense, and at times felt completely unmanageable. I remember the early mornings and late nights vividly: waking up at 5 a.m. to fix something, after going to bed at 3 a.m., feeling the weight of every little issue pressing on me. It was exhausting, but it was also exhilarating in a way the previous projects never were.
This startup started with an idea that felt right at my core. It was ambitious, technically challenging, and required a level of commitment I hadn’t experienced before. Compared to the previous ventures, this one demanded not just coding skill, but deep engagement with infrastructure, scaling, and constant problem-solving. Every day brought a new headache, a new obstacle to overcome — but the rewards were significant.
Looking back, this startup wasn’t just a project; it was a full immersion into what running a real, high-stakes venture feels like. It pushed me to my limits, taught me how to operate under extreme pressure, and gave me a sense of accomplishment that I hadn’t felt with the earlier projects. The journey was messy, chaotic, and sometimes brutal, but it’s also the startup I’m proudest of — the one that really showed me what I’m capable of.
This all started with a goal. Not some sudden spark of “let’s build a startup,” but a real problem I ran into during my first SaaS project: email.
Most email providers were expensive, limited, or clunky. I just wanted something simple, something I could run myself. So I built my own email server — with a Go API on top so I could access everything easily. At the time, I didn’t realize this would end up becoming the backbone of an entire startup.
The idea was simple: give people a service where they could add their own domain, set up custom addresses, and use it just like Gmail or Outlook, but cheaper and with more control. We even had a tiered model — free, basic, pro, and custom.
Sounds neat, right? In reality, it was overwhelming.
Nights blurred together. Sleep became optional. I’d crash into bed at 3 a.m., only to get pinged at 5 a.m. because a server had gone down or some massive bug had surfaced. Most nights, honestly, were sleepless. But those nights taught me how to break big problems into smaller pieces, how to keep moving forward even when the list of issues felt endless.
I even had to write my own SMTP server. On top of that, there were APIs, backend systems, and eventually mobile apps. I had to learn mobile development from scratch, just to keep the whole ecosystem alive.
There were four of us co-founders, grinding daily. When we opened the beta, we only wanted 500 testers. It filled up instantly. That rush was surreal. But reality hit quick — only about 20% converted to paid users, and most of those stuck to the Pro Plan. Not great strategically, but still validation that we’d built something people wanted.
We built cool stuff with it too. With just 30 lines of code, someone could spin up their own newsletter. That simplicity was the magic.
But the cost of keeping it alive was brutal. After five months of 17-hour workdays, we got an email: an offer to buy us out.
We sat down — four burned-out founders, staring at the fork in the road. Push further, risk breaking ourselves, or take the exit. We chose to exit. Not because the project wasn’t promising, but because we were running on fumes.
And that’s how it ended — not with a failure, but with a choice.
The Journey So Far
One of my ventures resulted in a modest exit—I sold it for approximately xxx amount, splitting the proceeds with my two co-founders. While not the transformative success I had initially dreamed of, it wasn't a failure either. It occupied that challenging middle ground that entrepreneurs rarely discuss—the "small win" that doesn't make headlines but still represents achievement.
What I Would Do Differently
1. Prioritize Mental Health as a Business Asset
This was perhaps my greatest oversight. I allowed the relentless startup grind to consume me entirely—stress, long hours, and constant pressure became my daily companions. Moving forward, I understand that my mental health isn't separate from business success but foundational to it. When my mind is clear and balanced, my decision-making improves, creativity flourishes, and resilience strengthens. Mental health isn't just self-care;
2. Treat Financial Management as a Core Function
Financial discipline wasn't our strong suit, particularly regarding operational expenses like cloud services. Our AWS costs spiraled as we scaled, something we could have prevented with better planning and optimization. In future ventures, I'll approach financial management with the same rigor I apply to product development. Cash flow, expense management, and resource allocation deserve focused attention from day one—not as afterthoughts when problems arise.
3. Validate Ideas Through Customer Interaction, Not Assumption
The most expensive mistake an entrepreneur can make is building something nobody wants. I've spent considerable time and resources developing features and products based on assumptions rather than evidence. Going forward, I'll validate ideas through customer interaction as early as possible. This means testing core hypotheses before significant investment, gathering meaningful feedback, and adapting quickly based on real-world responses.
4. Embrace "Good Enough" for Launch
The pursuit of perfection became a silent killer of progress. We repeatedly delayed launches, convinced that one more feature or refinement would make the difference. What I now understand is that a launched product generating real feedback is infinitely more valuable than a "perfect" product that remains hidden. The initial version doesn't need to be flawless—it needs to solve a core problem well enough that users see its potential value.
5. Recognize the Warning Signs of Diminishing Returns
One of the hardest entrepreneurial skills to develop is knowing when to pivot or stop. We persisted long after the evidence suggested we should change course, fueled by initial enthusiasm and unwillingness to face difficult truths. Extended periods of diminishing returns and increasing burnout should have triggered earlier reassessment. In future ventures, I'll establish clear metrics for evaluating progress and be more willing to make difficult decisions when those metrics indicate problems.
Fourth startup
The fourth startup is different. It hasn’t launched yet, so I can’t say too much. But it’s alive — there’s already a working version, and I’m planning to put it out soon.
This time, I’m building it alone. That changes everything. No co-founders to share the load, no one else to carry the weight when things get overwhelming. It means the pace is slower, but also that every decision — every line of code, every product choice — is mine.
I expect it will take two to three months before it’s really ready, but for the first time, I also have something I didn’t in the past: cash flow. That gives me space to think bigger, to shape this into more than just a side project gasping for air.
The plan is simple: start with a free service or a free trial, then grow from there. I’ve got Glass Pro lined up for hosting, and the foundation feels solid. Where it goes from here? I don’t know yet. But that’s the part I’m excited about — the uncertainty, the chance to see what happens when I build this one my way, from the ground up.
I believe in this idea. It feels universal — something almost anyone could use. I know I would. My friends agree too: they see the potential, even if it’s still rough around the edges. The truth is, it’s not refined yet. The UI feels off, the usability isn’t there, and I know the first impression matters.
So my goal right now is simple: get it to a point where it’s workable and looks good enough to stand on its own. Launch first, polish later.
It’s overwhelming, and because I’m doing this alone, progress is slower. But every penny I have is going into this startup. That’s how much I believe in it.
I haven’t seen anything quite like this out there. That’s what excites me — and scares me. Because if I’m right, it could be something bigger than just another project.