# The first time I started DevBatch I did not want to continue > Sixteen years ago I tried to start a company three times in twelve months. The first two failed. I want to write down what I actually thought during the third try, before the success made the failure look like a story. Source: https://waqaspitafi.com/writings/the-first-time-i-started-devbatch/ --- Published May 18, 2026 Reading 5 min Pillar Building durably Note Signature piece Share [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwaqaspitafi.com%2Fwritings%2Fthe-first-time-i-started-devbatch%2F) [X](https://x.com/intent/post?url=https%3A%2F%2Fwaqaspitafi.com%2Fwritings%2Fthe-first-time-i-started-devbatch%2F&text=The%20first%20time%20I%20started%20DevBatch%20I%20did%20not%20want%20to%20continue) [Essay](https://waqaspitafi.com/writings) # The first time I started DevBatch I did not want to continue Sixteen years ago I tried to start a company three times in twelve months. The first two failed. I want to write down what I actually thought during the third try, before the success made the failure look like a story. The first time I started DevBatch I did not want to continue. In late 2010 I had just been let go from Emirates Airline in Dubai. I was thirty-four. I had two young children and a wife. I had a UK skilled migration application that had been rejected on a clerical reading of my own bank statement. I had no visa to stay in the UAE and no job to keep one. I went back to Pakistan and accepted a fifty-fifty partnership offer from a friend who had a software services setup in Islamabad. I asked to move it to Lahore. He agreed. A year in, the partnership broke. The agreement said he would handle the money and bring the work; I would run the delivery. The reality was that the team was sitting on my head, the payments were coming out of my account, and the work was not arriving. He came down with a few friends who looked, to be honest, like they were not there to negotiate. We held the line on the partnership deed I had written and the partnership ended on the deed’s terms. I went back to Multan for two weeks. I had a car accident in this period. My car went under a truck. We were all saved. I had personal trouble in this period that does not belong on this site. I came back to Lahore and tried again. I rented a house large enough that the drying-room area could be a working space. I had one customer for one project. I had not written code in five or six years. I sat at the keyboard and wrote the software myself for two months. A friend named Hassan came in and asked if his textile-company team could share the space; he paid me rent. I hired my first developer, an iOS engineer named Asad. Then a man came into the office and proposed another fifty-fifty partnership. He was the sales side. I was the build side. We tried it. Three months later that one ended too. The third try, I started the company alone, picked the name DevBatch, and have not had another partnership since. I am writing this down because the version of this story you usually hear is the version where the third try is treated as the inevitable success and the first two are treated as preface. That is not how it felt. The third time I started DevBatch, I did not want to continue. I had run out of stories I could tell myself about who I was. I had been a programmer, a project manager, a Dubai expat, an Emirates employee, a husband, a father, an emigrant, and now twice a failed partner. None of those stories fit the man sitting in the drying-room of a rented house in Lahore writing his own code at thirty-five. The reason I continued was not heroic. It was that the alternative was worse. I had no visa. I had no job. I had a family. The only thing left to try was to do the same thing one more time, alone, and see if the failure mode was the partner or the founder. It turned out to be the partner. If I had to write down what I learned in those twelve months, I would write three things. The first is that a partnership deed is one of the most underrated documents in a small services business. The deed I wrote with my first partner did not save the partnership; it ended on its own terms. But the deed protected me when the conversation got rough. The cost of writing a deed is one weekend. The cost of not writing one is open-ended. Always write the deed. The second is that it is easier to build a company alone than to build it with the wrong partner. This is not the conventional advice. The conventional advice is “find a co-founder.” I tried that twice. Both times the partnership cost me more than the solo work did. I am not saying co-founders are a bad idea. I am saying the conventional bias toward two-founder teams underweights the cost of the wrong second founder. If you are not certain, default to one. The third is that the worst year of a company is the year before it has anything to show. In 2010 and 2011, every conversation about DevBatch was a conversation about a thing that did not exist yet. Every prospect call ended with “let’s stay in touch.” Every recruiting conversation ended with “let me think about it.” The company existed only inside my head. The reason most companies die in this period is not that they had a bad idea. It is that the founder runs out of permission. They stop giving themselves permission to keep pretending the company is real until it is real. The third year, the company was real. Customers paid. Engineers stayed. The phone rang for inbound. By 2013 I knew the work was going to compound. By 2017 we had paid four thousand monthly salaries, which I worked out at the time was roughly four thousand kitchens fed and twelve thousand people supported. It is 2026 now. The company is still mine. I am still the only owner. We are sixteen years in. The longest customer relationship is over five years. The healthcare-enterprise vertical that has grown out of long-running hospital network engagements is opening conversations with US hospital groups for the first time. None of that would have happened if the third try had ended the way the first two did. The thing I want operators reading this to take away is the smallest one: the years that look like a clean origin story in retrospect were not clean while they were happening. The third try was not the inevitable one. It was the one I happened to survive. Most of what I have built since then was built on top of having survived 2011 specifically, and not on top of any decision I made before that. If you are in the year nobody can see yet, that is the year. Do not stop. #building#services#devbatch#founding#durability Waqas Khan Pitafi Founder and CEO of DevBatch since 2010. Sixteen years building one engineering services firm, from Pakistan, now in Dallas. [Follow on LinkedIn](https://www.linkedin.com/in/pitaficodes) Next step ## If this was worth your time, there is more of it I read every message that comes through the form. Specific asks get faster replies than open-ended ones. [Start a conversation](https://waqaspitafi.com/connect) [Read more writing](https://waqaspitafi.com/writings) Keep reading 1. [ July 6, 2026 · 3 min ## What four thousand monthly salaries actually mean I am not an inventor. I am not building something the world has never seen. I run a services firm, and the way I think about its impact is through the salaries it has paid out, one month at a time, for sixteen years. ](https://waqaspitafi.com/writings/4000-kitchens) 2. [ June 22, 2026 · 3 min ## The uneducated engineer Three buckets of education, and the one we have stopped filling. Why I think the modern engineering profession is graduating brilliant people who are still uneducated. ](https://waqaspitafi.com/writings/uneducated-engineer)