The Dehydrated Company Hiring Model
Only hire when a function is underwater, so headcount forces ruthless prioritization
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 4
- Confidence
- 95%
A hiring discipline where you refuse to add a person to any function until the team working it is visibly 'underwater' — over capacity and asking for help. The company is treated like a dehydrated entity where each hire is a little water added only when genuinely parched. Slack in headcount is treated as a liability that manufactures low-value work and politics, not a buffer.
Origin
Varun Mohan's operating model at Windsurf/Codeium (~160 people, 50+ engineers), described as the way he and co-founder Douglas run headcount.
Core principles
- 01Idle capacity doesn't stay idle — well-intentioned people invent unimportant work to justify their role and then lobby to make it seem important
- 02You win by doing one thing exceptionally, not ten things adequately
- 03A high technical bar means you can't quickly spin up hires anyway, so restraint costs little
- 04A person's value is unrelated to the size of the team they own; owning people is a 'bad and gnarly idea'
How to run it
- 1
Anchor headcount to ambition, not to a lean-for-its-own-sake ideal
Ask the honest question: what is the minimum number of people needed to build the crazy ambition you actually have? Don't idolize being a tiny company; idolize being the smallest company that can satisfy your ambition.
Pro tip Sanity-check ambition vs team size out loud: '10 people building an autonomous vehicle' should sound un-serious. If it does, your headcount is wrong for the goal.
- 2
Hire for a role only when that function is underwater
Do not add to a function unless the people in it are effectively raising their hands saying 'we're dying, we need one more person.' Absence of that signal means no hire.
Watch out If you staff a function to comfort rather than to need, you import politics and invented work, not throughput.
- 3
Treat capacity pressure as a prioritization forcing function
When someone says they can't finish X in the time available, have the conversation: what can you do instead? Often the right answer is to deprioritize X rather than hire — using the constraint to clarify what actually matters.
Pro tip Founders also want to do everything; being forced to say 'we cannot do X' is clarifying for leadership too, not just for the team.
- 4
Keep people fluid across projects, never owned by a manager
Assign directly-responsible individuals to projects and move people to whatever project is most important. Reward the person who ships the craziest project with the fewest people.
Watch out Tying a leader's status to team size incentivizes empire-building, which is exactly the failure mode this model exists to prevent.
In the wild
Mohan's rule: unless the team building inference technology is underwater, they will not hire someone to work on it — even though it's core to the product.
→ Headcount stays tied to genuine need; trusted, high-bar people never ask for work that isn't important, so requests for help are trustworthy signals.
Common mistakes
Adding a buffer 'just in case'
Even 20% surplus headcount undermines the ruthless prioritization the founders are trying to enforce; the extra people will find and defend low-value work.
Idolizing leanness as the goal
The point isn't to stay a 10-person company forever — under-hiring for a genuinely large ambition guarantees failure just as over-hiring does.
Is it for you?
Best for
Founders and leaders of high-bar, fast-moving startups who trust their people and want headcount to enforce focus
Not ideal for
Organizations with lower hiring bars, low mutual trust, or predictable/commoditized work where slack capacity is genuinely useful
From the transcript
“I wanted the company to almost be like this dehydrated entity right and every hire is is like a little bit of water, right? And…”
“we only hire for a role if we're actually underwater for that function”
“You don't win by doing, you know, 10 things kind of well. You win by doing like one thing really well and maybe you fail…”
“the person that is the most valuable at the company is the person that can do the most crazy sort of project out there with…”
From the episode
Building a magical AI code editor used by over 1 million developers in four months: The untold story of Windsurf
Varun Mohan (co-founder and CEO)