11 Months at the Triangle Factory
A few months ago, someone asked me what my job title was. "Chief of Staff, Product & Engineering," I said. They paused. "Okay, but what do you do?"
It's a fair question. I've spent the last eleven months at Vercel doing three jobs that shouldn't be one job: Chief of Staff, leading ShipOps, and something I've started calling a Forward-Deployed Executive. They overlap constantly, and the overlaps are where most of the value lives.
I want to make the case that doing these together is more impactful than doing any of them alone. But I also want to be honest about why that's true, because it has less to do with me and more to do with the kind of organization Vercel is.
The water I swim in
In 30 Days at the Triangle Factory, I tried to describe Vercel:
It's hands down the fastest-moving, fastest-shipping, highest trust/autonomy company I've ever seen with an incredible density of talent. And at Vercel — this works, but I don't think it would at most other places.
Eleven months later, that description still holds. We don't try to solve everything in our heads. We ship something that's 10–20% done (internally), get feedback, and iterate. Work happens in public Slack channels. Decisions, drafts, and debates are all visible, which reduces the game of telephone and lets context travel faster than formal updates ever could.
What's striking isn't any single moment; it's the texture of the system. You feel speed less in big launches and more in the absence of drag. Work doesn't disappear into private docs. People from other teams routinely drop into channels because they're paying attention and the context is there.
This way of working isn't clean. It requires emotional resilience and comfort with ambiguity. But it creates a learning velocity that's hard to replicate otherwise. And it means that when things need coordination, they need it fast, which is where my role lives.
What I actually do
Chief of Staff
When I joined, my mandate was to build the operating system for the Engineering, Product, and Design (EPD) organization. Chief of Staff roles are highly varied (even at Vercel, the three of us have materially different mandates). Mine boils down to two things: improve velocity and alignment for EPD and provide leverage to the CPO and CTO. The obvious activities are here: running weekly staff meetings, ghostwriting presentations, putting together board slides, and acting as the CPO's proxy in different venues.
But the part of this work I've come to value most is one I didn't anticipate: acting as a staging environment for ideas between leadership and the organization.
A regular interaction.
Someone on the leadership team has a half-formed thought about how a team should be structured or a new ritual they want to start. They bring it to me. I ask questions, reframe it, poke at the assumptions, and help them figure out whether it's ready to share, and if so, how. Sometimes the idea lands intact. Sometimes I help them realize it needs another lap. And sometimes the best thing I can do is tell them the idea is good but the timing is wrong, that the organization is digesting something else right now and this would land poorly.
I have a degree in rhetoric and I'm used to reframing conversations with metaphors. I'm better than average at asking thoughtful questions thanks to a decade in user research. Those things are necessary, but they're not sufficient. The thing that actually earns me the right to play this role is that people trust me. They know and feel that I care about Vercel, both the company and the organization. The first two make me good at the work. The third is what gets me invited into it.
ShipOps
ShipOps evolved out of our technical program managers with a simple goal: help the company ship faster and stay aligned as it scales. We talk about our work like a Formula 1 team: our goal is to enable everyone to focus on what they do best and let us handle everything else.
In practice, ShipOps is the social and technical infrastructure that keeps the organization functioning. We own the rhythm of the business, including: company planning, monthly business reviews, twice-a-week product reviews, the Launch Calendar, core EPD pages in Notion, and Slack channel architecture decisions. We don't tell teams what to build. We focus on how work moves.
One strength of the team is that both TPMs were engineers in their prior lives. So we don't have a hammer of "process" and run around thinking everything is a nail. Our principles are explicitly the opposite:
Minimum necessary process — long-term leverage and guardrails over rigid recipes.
Normalize it, then formalize it — start with an MVP and ITG* based on what's working. Progress > polish (usually).
Reduce risk, complexity, and comms barriers — get rid of the things in your way, keeping the best interests of the business in mind.
*ITG is "iterate to great" in Vercelian.
Forward-Deployed Executive
Because of my two official roles, a third one has emerged, and it's the least legible and most interesting part of the job.
A forward-deployed executive is a professional fixer. You're given (or find) problems, not projects. You go where context is missing, ambiguity is high, or ownership is fuzzy. You do whatever needs doing to get something into a stable state, and then you hand it off or fade into the background.
In summer of 2025, we launched a new monthly business review process. I wasn't the owner, but I was deeply involved: chief document editor, primary sheep herder, the person following up on action items and next steps. A few months in, when we started discussing changes and I took on more responsibility, people were surprised to learn it hadn't been "mine" all along. We've since brought in a new owner, and my role has shifted again to supporting her as she makes it her own.
That's the job. If you're doing it well, ownership becomes blurry because the work is about quality and outcomes, not credit. I've come to think of this as a distinct shape of work, not a role on an org chart. You embed, you diagnose, you restore clarity, and you step back.
The tension
Traditionally, these three roles would be distinct, often by necessity. Organizations create clear ownership, defined scopes, and explicit authority because that's how they manage complexity. A Chief of Staff has a lane. A TPM org has a lane. And "professional fixer" isn't a lane at all, which is why it usually doesn't survive contact with an org chart.
I sometimes joke that I'm a high-paid janitor because people give me shit to clean up (lovingly). Most of my impact across all three roles is invisible. The roles themselves are forms of infrastructure, and you only really notice infrastructure when it's missing or it breaks. When people around you are executing with more clarity and less friction, something is working — even if no one can quite point to you as the reason.
That's the tradeoff. The work matters most when it's hardest to see.
What holds this together
If the three roles are the what, care is the why. Not care as a soft abstraction, but care as a function.
I was not hired to build processes. I was hired to care. That means diagnosing where decisions are stalling, where ownership is unclear, and where coordination is breaking down, then building whatever is needed to fix it (processes, rituals, agents, etc.). This is why my team is not goaled on the health of our programs but on the effectiveness of the organization. The programs are simply one of many vehicles to deliver value, and they will be as short or long-lived as necessary.
This only works because of trust, and trust in a few different directions. Trust from leadership that I'll use my autonomy well. Trust earned across the organization by not empire-building or competing for credit. And trust in judgment and competence, which is a different thing from trust in intentions.
I also think my work has been impactful because I sit at the intersection of "tremendous context" and "open space on my calendar." This is different from the leadership team, who has maximum context and minimum open space. Companies have so much potential energy trapped in meeting notes and to-do lists. My job is to ensure that energy becomes kinetic.
Less process, more discipline
I've always been fascinated with how people work together, and I've spent a good chunk of the last decade thinking about process. Being at Vercel has changed some of what I believe.
I used to think processes needed to be owned and maintained indefinitely by the people who created them. Now I think of processes as modular, more like software. If you're clear about why a process exists, what problem it's solving, and how it works, then teams can fork it, extend it, and make it their own. Ownership doesn't mean central control forever.
I still believe that process raises the floor more than it impacts the ceiling. But being at Vercel has taught me something counterintuitive: less process requires more discipline. The real bottleneck in any organization isn't process. It's trust. And trust is harder to build and maintain than any process ever will be. This isn't unique to Vercel — every growing organization I talk to is navigating some version of this tension. Which brings me to what that actually looks like in practice.
A good example is our Launch Calendar, where we keep track of everything the company is shipping. While ShipOps "owns" Launch Calendar, a number of teams have built their own views and workflows on top of it to help them do their work, whether that's custom automations or just making it part of their weekly cadence. Instead of being prescriptive about how everyone has to use Launch Calendar, we are clear about what Launch Calendar is (and is not) and what's expected of the organization. From there we trust and support other teams to use it as they need.
And while we own one way to create entries, we don't block other teams from building their own agents to do it, because being clear about expectations lets us be flexible about implementation.
(Un)Learning, continued
In "30 Days," I wrote about how Vercel was challenging a decade of corporate culture and nudging it out of me. On Day 12, I noted that "Vercel felt uncomfortably fast at first, now it just feels fast." Eleven months in, that process hasn't stopped. It's just shifted from the pace of the work to the shape of it.
Something I didn't appreciate enough coming in is how much my time in consulting set me up for this role. As a consultant, I knew my time was limited, and my primary job was to be leverage and make someone else successful. The same is true here. There is no such thing as a solo victory in this kind of role. The difference is that inside the company, you get to see the compounding effects of that leverage in ways you never could from the outside, which is something I noted in "30 Days" but have now lived long enough to confirm.
I still think that starting the #behzod-notes channel and posting daily observations and reflections is one of the most valuable things I've done for myself and the company. Across the first two months, I talked to about 20 people each week from across the organization, and I've now shared about 170 posts of varying length. The discipline of observing and writing daily has become central to how I think and work. This experience has reinforced that curiosity, far more than judgment, is the key to any success I'll have here — which is increasingly true as we continue to navigate a changing work environment and company growth.
Looking forward
Six months from now, I'd love to say that ShipOps feels like a team of leverage engineers, not just running programs and processes, but building more tools, systems, and agents that help the company operate more efficiently, often without us in the loop.
As organizations continue to change faster, it's useful to have someone turning over rocks and questioning everything, assuming that person can do something with the answers. A person who can come in with no ego and roll up their sleeves and do what the organization needs (not just apply a playbook) is a powerful asset. As I talk to people at organizations where work is changing at an accelerating rate, I'm increasingly convinced that more companies will need to bring in people who are simply hired to care.
Care, leverage, and infrastructure aren't particularly shiny. But in the right environment, they compound.
Thank you to Arthur Nelson, Brian Dell, Crystal Widjaja, David Choe, Elan Miller, Michael Nguyen, Radhika Bhalla, Sam Spurlin, Tina He, and Tom Critchlow for your feedback on prior drafts of these ideas.