Darren Murph is the "oracle of remote work", according to CNBC. He scaled Gitlab's fully remote operational model from startup phase through to the brand’s historic IPO, and now runs his own advisory and consulting firm. Here's the framework he uses to help founders build a distributed model that scales.
In 2019, Darren Murph received a career-defining phone call, from a software startup with no offices anywhere in the world. “I remember they told me they wanted to do the impossible: they wanted to become the first fully distributed company to ever go public in the United States.”
Hoping - perhaps, foreseeing - that this was an opportunity to make history, he said yes and became GitLab’s first-ever Head of Remote. “It was extremely rare to find a proper, intentional, all-the-way remote work environment at the time. And I thought, if we do this right, the world will pay attention.”
Murph joined GitLab when the team was made up of around 500 people. By the time it IPO’d in 2021, it had scaled to over 2,000 across 65 countries. When Covid-19 hit, the remote-first playbook Murph had created for GitLab became the blueprint for the entire world.
During that time, and throughout the course of his 20-year remote-based career, he’s seen time and time again the stark difference between companies that start out distributed on purpose, and those that pivot once they’re already established. Spoiler: Those that go in with intention are way more likely to achieve longevity - and have an easier time scaling overall.
What a startup gains by going remote-first
Right away, a remote-first startup gains considerably more runway compared to hybrid or entirely in-office organisations, says Murph. “Imagine you’ve just received your Series A cheque, and then immediately have to spend up to 20% of it on an office. If you’re fully remote, that money goes into research, development, hiring - things that actually matter in your business.”
The second gain is what Murph calls ‘speed off the line’: “Imagine you’re running a race at the Olympics. When the gun goes off, you want to launch off the line as quickly as you can. It’s the same with your business. In-office startups waste the first crucial moments finalising lease admin and choosing the right ergonomic chairs. Meanwhile, the remote startup is starting work on day one - launching right off the line.”
The third major gain is talent density - and this is by far the biggest win, according to Murph. “You can literally hire the world's best people, because you don't have to convince them to move somewhere else. The location-agnostic role can be a compelling counter-offer for potential hires, who are weighing up packages from big names like Nvidia or Google.”
How to go remote from day one: 3 area of operation to get right
Before anything else, the most important thing for a founder to do is to commit fully, says Murph. “Whatever you choose, it shouldn’t be implemented as an experiment - it has to be positioned as the default, so that you can purposefully build the infrastructure and processes needed to make it run smoothly.”
Changing the approach a couple of months or years in runs the risk of activating a deprival super reaction across the entire organisation. Murph explains: “When you give something to someone when they start working with you, they accept it as a default. But if sometime later you significantly change it or take it away from them, they will super react to the perceived loss. Which means a reaction that's past the point of a conversation, it’s impossible to control.”
Now, imagine that reaction at the scale of hundreds or thousands of people. “This is the thing a lot of founders don’t understand: the power of human psychology. The thing that you're bottling up is only going to be more intense later on.”
Once the intention is set, says Murph, there are three core areas of operation that are critical to get right first: Culture, operational discipline, and hiring practices.
“Most founders think they can wait until they get to 50 people to put this together. But if you do that, you’ve got 50 people each with vastly different ideas about culture and operational discipline. Trying to untangle that is going to be impossible. The complexity only compounds with scale.”
A founder who works to get all three right from the start sets themselves up for success, says Murph. “Because it’s so easy to get right early and it’s so brutal to deal with even a few years in. It has to be the first thing you do, or you’ll spend the rest of your existence as a company wrangling dysfunction.”
1. Culture
For Murph, culture always comes first - because it’s one of the hardest things to implement once an organisation has already started scaling. “When you’re building culture for four people, it's not that complicated. If you’ve already hired 400 people and now you're trying to inject the culture into them - you’re going to struggle.”
He highlights two non-negotiable remote-first culture principles to get right early on:
- Transparency: “You have to be comfortable sharing things publicly and in the open, because that's the only way people learn via osmosis. Essentially, you need to treat your Slack channel like the open-plan office environment. You can't communicate in direct messages, you have to communicate in public.”
- Iteration: “When you're highly distributed, you're not going to be able to work on a typical synchronous schedule. It's going to be mostly async by design. So, your team needs to be comfortable with sharing work in progress and moving things forward in small components - which links back to transparency, because they have to do it out in the open.”
Importantly, remote-first doesn't mean remote-only. Murph urges a reframe against the binary all-or-nothing thinking, with a reminder that the core principle of a remote setup is flexibility, not never-ever seeing each other in person.
“In the early stages, it’s probably worth making a point to have some sort of cadence around getting together on a regular basis. It might be once a month, before you move to quarterly or twice a year. Maybe only a certain subset of people meet regularly. It's going to be different depending on the startup.”
Regardless, the in-person meetups that do take place absolutely need to be worth the travel. “You have to curate an experience, to plan moments that are going to be extraordinary. You may even end up hiring someone specifically to craft these,” says Murph.
2. Operational discipline
In a traditional non-distributed startup, processes and operational procedures can often arise organically as needed. They tend to get transmitted throughout the team and to new hires almost via word-of-mouth. In a remote-first environment, however, organic transmission doesn’t have a physical stage on which to take place.
Embedding a culture of documentation is therefore critical. “If you have to wait 10 hours for someone to wake up and have a synchronous call, you're going to be really, really slow. If you codify the work that you did by writing it down or recording video, you can create a kind of relay system - an always-on type model.”
Growing an org brain
To support this always-on model, Murph developed the concept of an ‘org brain’ - a centralised place for the collective intelligence of the entire organisation. Essentially, it functions as a foundational system to enable collaboration across teams and timezones.
“In a distributed setting, you’re more likely to interface with a person's information stream a hundred times before you interface with them as a human,” says Murph. The org brain mandates that people record information rigorously and clearly, to enable that collaboration.
To build an org brain, Murph recommends starting with centralising all the information about how the company works, what tools are used, what to do when there's issues etc. “All of it has to be written down. This used to be tacit knowledge in an office, but now it has to be explicit knowledge.”
How to entrench a culture of documentation
So, how can a founder ensure that writing things down becomes second nature for their team? For Murph, the answer is part people, part tooling.
The biggest point of leverage, says Murph, is to select tools that are going to organically shape operational discipline as a byproduct of their everyday use. The more seamlessly these tools are integrated, the more difficult it's going to be for someone to get it wrong
“For example, using video tools that automatically record every non-sensitive meeting, and pairing that with an AI element that extracts and organises the outputs into the relevant knowledgebase that feeds your org brain. If this all happens by default, it will manage itself to some degree.”
For the parts tooling alone doesn't handle, Murph recommends hiring someone specifically to orchestrate all the systems that are implemented - a job that doesn’t tend to exist in in-office environments. “You need someone fully in charge of making sure the brain is running smoothly.”
Org brain and AI: The added benefit
For Murph, establishing a robust org brain has an added, urgent benefit in the AI era: It sets up a centralised knowledgebase for platforms and agents to draw from.
“Every company in the world is trying to layer AI into their workflows, and one of the first things you have to do during setup is connect to sources. But if you don't have a centralised knowledgebase, in Notion or Confluence for example, it’s going to be massively hamstrung.”
Without a comprehensive knowledgebase to draw from, AI will do worse than just return empty results - it will try and come up with answers, and the quality will be relatively poor, because it’s only as strong as the amount of data that it has access to.
“And that AI is going to be used hundreds or thousands of times per day. And so the cost of getting this wrong is extraordinary. You're going to spend so much time and effort revising and telling it why it's wrong,” says Murph.
Like it or not, says Murph, the ROI on AI is going to be much higher with comprehensive documentation and information in place. “Literally all of this can be solved by being intentional about knowledge as a product. Because now not only are your team using it, but the AI is using it. So, before you go hard on AI, you first have to step back and say, ‘Do we have an org brain in place? And do we have someone that runs information as a product?’”
Operational discipline loops back into both culture and hiring: “once the operational discipline is defined, you’ll find you’ve ended up writing down what it’s like to work at your organisation.”
Essentially, as a byproduct of defining ways of working, an organisation ends up with a document that they can attach to their hiring guide - outlining not just values, but actual things the business does that defines how work needs to be done, which gives potential hires a strong sense of what to expect.
3. Hiring
When it comes to putting a team together, Murph’s guiding question is: “Who is the type of person we want to hire that will reinforce our operational discipline and our culture?”
Over the course of his career, he’s seen time and time again that when a founder is very explicit and deliberate about the kind of company they want to build, they end up finding people that are like-minded and want to invest in building together.
Murph’s remote-first hiring checklist includes:
- People with high amounts of agency and autonomy: “They tend to thrive in distributed settings and to be high performers. A specific question you can ask is whether they want to run really fast with as little bureaucracy and as few meetings as possible. You want somebody to say yes to that very enthusiastically.”
- People with low levels of shame: “If you’re building a culture based on transparency, and asking people to take an iterative approach to working - then you need people who are okay with being vulnerable. If they aren’t, they’re going to actively work against your culture and your operational discipline.”
Checking both of these boxes is likely to land an organisation with a team of people who will fit the culture and adhere to the operational discipline. More importantly, Murph says, “they’ll want to level it up with you and push you to improve processes or switch to better tools for documentation.”
Get this right enough times, and a founder ends up with the critical mass needed to maintain culture as the organisation scales. “As long as the majority of people are invested in adhering to the operational discipline, then their effort will win over the people that are just mailing it in - the culture will stick, and everything just continues to work.”
Final thoughts
Before Gitlab, no one had run the remote-first experiment at such an extraordinary scale. Now, thanks to Murph and his team, the blueprint exists - and it’s helping smaller firms like Lovie and LEAD3R, as well as global powerhouses like Amgen and Capital One.
For founders weighing up their options, this is the core takeaway. Going remote-first is entirely possible, and can be hugely successful - provided it's baked into culture and processes with intention. And, the earlier an organisation commits, the less there is to retroactively build in later.
