I distinctly remember an 'oh no' moment during my first meeting at Render in December 2020. The entire company had gathered on Zoom to welcome me—all 14 of us, fitting neatly on a single screen. Render was still a seed-stage company. Just three weeks earlier, I had been in an all-hands meeting at Stripe, surrounded by thousands of colleagues. At that moment, the scale of my transition went from theoretical to very real.

This post is long overdue since I left Render at the end of April 2024 after seeing it through series A and B and 14 to 70 people. During my sabbatical, I reflected on my time at Render, but it felt too intimidating to compress 3.5 years of whirlwind startup life into a single post. However, I am starting a new role in January, so I need to get my thoughts down before I forget and get absorbed in a new job. Hopefully, this will capture some of the lessons and challenges of my experience.

Why Render?

I have answered this question so many times over the last few years that it feels pat and memorized. The more interesting part is the subtext behind the question: " Why leave the Stripe rocketship to take a risk on an early, unproven startup?”

I started chatting with Anurag (Render’s founder and CEO) in the fall of 2020. At this point, I wasn’t even thinking about leaving Stripe [1]. I was growing and learning and saw myself continuing to do that for many years. A lot was happening in the world with a pandemic, civil unrest, a critical US presidential election, etc. We were still in the worst parts of the pandemic, but from the inside, Stripe was booming and responding to the challenge of the entire world moving online rapidly. However, something about everything happening in the world made many people, including me, look at life goals differently. When our world was collectively turned upside down, it gave me a sense of urgency that if I didn’t pursue things I wanted to do, there might never be a good time. I had spent many years pursuing relatively safe career choices. I was yearning to do something different. Stripe subliminally contributed to this because the culture and people around me made me think much more broadly. A Stripe value was “Elevate Ambitions,” and they had undoubtedly elevated mine. I was now looking at options I would have previously written off. Render was also the rare startup aligned almost precisely with what I was a good fit for. They needed my infrastructure experience, and I was drawn to the exceedingly ambitious idea and highly talented team that Anurag had assembled. I knew I would regret passing on the alignment of people, team, and mission. I had to admit that it was a much riskier choice financially and professionally. Mapping out various scenarios, I had put a strong likelihood of either the startup failing or me failing in the role. However, I couldn’t dissuade myself and joined in Dec 2020 [2].

Each of the following sections can be its own post or series of posts, and I might get to them eventually, but this is a summary of the big buckets of my experience and learning at Render.

Learning the business side

My biggest goal was to learn all the ins and outs of a business, and I got to do this in even more detail than I had imagined. When I joined Render, no one exclusively supported HR, recruiting, finance, or similar functions. As the most senior leader after the CEO, I supported him or was often directly responsible for many things as they came up. Some of these were more fun than others, e.g., helping with the series A and B fundraising decks and board memos. Less fun was dealing with compliance-related paperwork, contracts, or individual offer letters that Anurag or I still filled in for all new employees until we hired our first recruiter. I saw firsthand how difficult marketing, sales, distribution, and organizing people are. I am a far better engineering leader due to learning and observing how various functions work closely.

Building software end-to-end

In addition to learning the business, I wanted to learn about the entire cycle of shipping software, and doing greenfield work was exciting. Some highlights were

Render was going through the classic early stage, where it went from a very solid product for a specific cohort of users (Silicon Valley VC-funded startups) to one that needed to scale to be a platform for a broader segment (small and medium businesses, agencies, and teams at larger tech companies, amongst others). We built the engineering organization to reflect this change, going from one team where everyone worked in a project-based model to multiple teams with their own charters that focused on differentiated user needs. In parallel, we had to build functions that enabled engineering, such as support and go from hiring generalists to specialists.
The competitive landscape was changing rapidly, with newer players joining and incumbents such as Heroku deciding to end their “free tier,” significantly impacting Render’s user growth. Right after the Heroku announcement, the entire engineering team spent weeks dealing with all the new users signing up overnight after the acquisition and scaling the platform to keep up with growth.
As Render grew, we had to reposition, reprice, and repackage the product and offerings to appeal to newer user cohorts. This meant many large projects, such as rewriting billing infrastructure, building a compliance framework to appeal to more mature users, spinning up many new cloud regions, etc.

Leading through a quickly evolving landscape

A key difference between being at a startup versus a larger company during challenging times is that at a startup, you look around and realize that you are one of the grown-ups in the room who will have to help navigate the upcoming storm. It’s definitely an “oh shit” moment the first time it happens, but you get used to it quickly.

2021 through 2024 feel like years where a lot happened very quickly. In some ways, those years felt like extensions of 2020 and the pandemic and its related upheaval. There was an extended tech boom through 2020 and 2021, followed by an almost immediate collapse with many ripple effects across all tech beginning in 2022. We all collectively got weary of using the word “unprecedented.”

A few key events I remember as being challenging were.

The end of ZIRP and the changing tech macroeconomic climate in early 2022 impacted everything from fundraising to customer growth to hiring. It’s hard to overstate how much that changed the entire tech and Silicon Valley climate.
The effects of the upheaval we saw in banking, particularly with the Silicon Valley Bank and First Republic Bank getting taken over, felt uncomfortably close. I will never forget the surreal experience of meeting with the rest of the executive team, talking through those events and our plan for them, and watching the Twitter meltdown happen, knowing that the implications for the tech ecosystem were quite unclear. I was very grateful when it was all done for a very efficient government system that was set up precisely for crises like this.
Github copilot was launched in 2021, and we paid attention to it at Render, but it was overshadowed by the biggest tech news of 2022, i.e., chatGPT, which is clearly the most disruptive event in tech this decade. We still don’t know the full ramifications, and I won’t attempt to make any predictions, especially in this blog post, but AI will clearly impact how developers work. At Render, we had to try to make calculated guesses on how we wanted to position ourselves, change our roadmap, and otherwise incorporate AI. This feels like a work in progress for the industry, and there are no clear answers yet.

Owning engineering means owning all the gritty parts of it as well

When you work on engineering or developer teams at large companies, teams get very specialized, and the less fun parts have dedicated teams dealing with them. However, at a startup, everything, including IT, Security, users engaging in fraud and unsavory activities, and many other random activities, all fall under the broader engineering umbrella. If you provide a service on the internet, there will always be highly motivated users who will find a way to use it in ways that were not intended and are often illegal. This is particularly true when a product is still being built; users know to test it out in ways it isn’t tested for. Building version 0 of Security, Compliance, and Fraud detection can feel like a game of whack-a-mole.

Hiring and building teams

I had substantial hiring experience before joining Render because Stripe expected all hiring managers to be very hands-on in all areas. I had worked with some fantastic recruiting partners in a rapid growth era (my team, for e.g., went from 12 to 40+). However, hiring at an early-stage startup in a tight labor market is completely different and requires unique methods and skills (the market then changed almost overnight). I learned about a few different flavors of hiring: hiring internationally, hiring recruiters, hiring contractors, hiring for many first-in roles, hiring support engineers, hiring security engineers, hiring peer executives, and designing compensation and performance bands. Building and maintaining a high-performance team is also very different at a startup, where you don’t have access to the same structures and support as you do for performance management, compensation reviews, etc. Importantly, there is also much less time and leeway and more urgency.

Evolution of my leadership style

When I interviewed at Render, Anurag asked me if I had a “working with Uma” doc. I didn’t at the time, but I wrote one up. Rereading it recently, I noticed that while integrity and transparency remain core to my leadership style, other aspects have evolved significantly. For instance, I wrote that I didn’t enjoy being surprised—a sentiment shaped by my years in structured environments. After navigating the unpredictable landscape of startup life during the aforementioned “unprecedented times”, I’ve developed stronger antibodies to surprises.

A section I didn’t talk about in the original document but one I have grown the most in is I am a far more decisive person at work now. While I’ve always been decisive in my personal life, I initially leaned on a consensus-driven management style at work.

At a startup with a small team, everyone, including individual contributors, is always making critical decisions. As unresolved decisions inevitably flow upward, I’ve learned to make quicker, more effective decisions.

This shift has also made me question the effectiveness of 'servant leadership,' which I now believe can sometimes hinder both leaders and their teams. Good leadership requires being deeply involved in the details, asking hard questions, and holding people accountable. Avoiding these responsibilities is, in essence, an abdication of leadership. I can write multiple posts on this topic, which overlaps with the viral “founder mode” essay in some ways [3].

There has been a lot of chatter in tech about “wartime vs. peacetime leadership,” but in my view, good leadership and what it means have not fundamentally changed. What has changed is the slack that allowed incompetent people, both leaders and others, to survive in the ZIRP era.

Constantly giving away Legos

With every process, team structure, or similar I built, I had to be conscious of building it to scale and be eventually handed off. This was the only way both I and the company could scale. The tricky part was knowing ahead of time when to invest in the process for scaling and growth vs when to treat it as a one-time effort. I didn’t always succeed, and many areas saw false starts where I delegated too quickly or held on to things I owned for longer than I should have.

My biggest realization about delegation came when I decided to leave and had to make decisions about things I owned. I now recommend that senior leaders do an exercise in succession planning every year to force the actual planning and consider what they would delegate if forced to.

People

In trying not to be “cringe,” as the kids say, the previous section of this post has taken on an overly businesslike tone, but I’ll allow myself this section. The best thing about Render is the people. All my favorite jobs have been about the team, and Render was a peak people experience. It was a group of some of the smartest, kindest people I have worked with, and I am very proud of everything we have accomplished. We spent hours together as a team, sometimes in hours-long highly critical incidents, sometimes on Zoom happy hours during COVID, and I have many “core memories.” I won’t try to name everyone, but hopefully, you all know who you are. Leaving the team was probably the hardest part of leaving the role.

As I alluded to earlier, I joined Render for the people, mission, and learning, but had tried to temper my expectations on outcomes. The experience exceeded my expectations, and I am proud of everything we were able to do as a team. It has made me much more well-rounded as an engineering leader, and I look forward to applying my lessons in future roles.

[1] Funnily enough, he reached out to me via Twitter DMs and is one of the reasons I found old Twitter so amazing and mourned its demise

[2] I sometimes use a decision-making framework to ask myself what I wish I had done when I was 70. Between Stripe and Render, it felt that there were no wrong choices, but 70-year-old me would regret passing on the opportunity to work at a startup with a mission, people, and values that strongly resonated

[3]I disagree with the way the essay structures its arguments. Imo, Cedric Chin and Charity Majors wrote the best counters to the essay https://paulgraham.com/foundermode.html

https://x.com/ejames_c/status/1830409563168034827

https://charity.wtf/2024/12/17/founder-mode-and-the-art-of-mythmaking/

Thank you to everyone at Render for contributing to my time there, particularly Anurag, for the opportunity.

Questions/Feedback/Thoughts? Email me at email [@] uma.dev