A client called me in because their production servers kept crashing. Every few days, seemingly at random, their EC2 instances would just die. When I asked to see their logs to understand what was happening, they had to SSH into individual servers and grep through files. That’s when I knew we had a bigger problem than just crashes. Their servers were running out of disk space. Not from user data or database growth, but from their own application logs. Every time their application did anything, it wrote to local disk.

A client called me in because their production servers kept crashing. Every few days, seemingly at random, their EC2 instances would just die. When I asked to see their logs to understand what was happening, they had to SSH into individual servers and grep through files. That’s when I knew we had a bigger problem than just crashes. Their servers were running out of disk space. Not from user data or database growth, but from their own application logs. Every time their application did anything, it wrote to local disk.

A client called me in because their production servers kept crashing. Every few days, seemingly at random, their EC2 instances would just die. When I asked to see their logs to understand what was happening, they had to SSH into individual servers and grep through files. That’s when I knew we had a bigger problem than just crashes. Their servers were running out of disk space. Not from user data or database growth, but from their own application logs. Every time their application did anything, it wrote to local disk.

I keep having the same conversation with founders. They want to discuss AI adoption strategy. I ask them why they think they need AI.

The silence that follows tells me everything.

Most Series A companies don’t need AI. They think they do because investors are asking about it, because competitors are claiming AI features, because it feels like falling behind not to have an AI strategy.

But need and want are different things. And premature AI adoption is expensive in ways that aren’t obvious until the damage is done.

I keep having the same conversation with founders. They want to discuss AI adoption strategy. I ask them why they think they need AI.

The silence that follows tells me everything.

Most Series A companies don’t need AI. They think they do because investors are asking about it, because competitors are claiming AI features, because it feels like falling behind not to have an AI strategy.

But need and want are different things. And premature AI adoption is expensive in ways that aren’t obvious until the damage is done.

I keep having the same conversation with founders. They want to discuss AI adoption strategy. I ask them why they think they need AI.

The silence that follows tells me everything.

Most Series A companies don’t need AI. They think they do because investors are asking about it, because competitors are claiming AI features, because it feels like falling behind not to have an AI strategy.

But need and want are different things. And premature AI adoption is expensive in ways that aren’t obvious until the damage is done.

I prepared technical due diligence materials for Blockdaemon’s Series A, B, and C funding rounds. Three times I had to answer the question: how do we know your technology actually works at scale?

The answer that got us to $3.25 billion valuation wasn’t the one most CTOs think matters.

VCs don’t care about your velocity metrics. They don’t care about story points. They barely care about your deployment frequency.

What impressed them was simple: we were running infrastructure for Meta and JP Morgan across five continents, and it worked.

After Blockdaemon hit $3.25 billion valuation, I started talking to companies about permanent CTO roles.

None of them felt as challenging as what I’d just done. Scaling from 10 to 187 engineers. Building infrastructure that Meta and JP Morgan trusted. Deploying validators across five continents. Managing 96% retention through hypergrowth.

I could wait for another opportunity like Blockdaemon. But I’d probably wait forever.

So I went fractional instead.

Why fractional made sense

The decision wasn’t complicated. I’d spent years building hard-earned knowledge about scaling technical teams, infrastructure, and organisations. I could apply that knowledge to help multiple companies rather than waiting for one perfect opportunity.

I keep having the same conversation with founders. They want to discuss AI adoption strategy. I ask them why they think they need AI.

The silence that follows tells me everything.

Most Series A companies don’t need AI. They think they do because investors are asking about it, because competitors are claiming AI features, because it feels like falling behind not to have an AI strategy.

But need and want are different things. And premature AI adoption is expensive in ways that aren’t obvious until the damage is done.

New data from LeadDev shows 38% of engineering leaders report less motivated teams than a year ago. Only 21% describe their teams as healthy. 12% say they’re emotionally drained daily.

If you’re a founder or engineering leader reading those numbers and thinking “my team seems fine,” you might be six months behind reality.

I’m seeing this across my network right now. Good engineers going quiet in standups. High performers suddenly interviewing elsewhere. Teams that used to ship fast now dragging on simple tasks.

New data from LeadDev shows 38% of engineering leaders report less motivated teams than a year ago. Only 21% describe their teams as healthy. 12% say they’re emotionally drained daily.

If you’re a founder or engineering leader reading those numbers and thinking “my team seems fine,” you might be six months behind reality.

I’m seeing this across my network right now. Good engineers going quiet in standups. High performers suddenly interviewing elsewhere. Teams that used to ship fast now dragging on simple tasks.

New data from LeadDev shows 38% of engineering leaders report less motivated teams than a year ago. Only 21% describe their teams as healthy. 12% say they’re emotionally drained daily.

If you’re a founder or engineering leader reading those numbers and thinking “my team seems fine,” you might be six months behind reality.

I’m seeing this across my network right now. Good engineers going quiet in standups. High performers suddenly interviewing elsewhere. Teams that used to ship fast now dragging on simple tasks.

When I told our CEO I wanted our best blockchain engineers to work part-time for Ethereum Foundation and Solana, he didn’t blink. Constantine trusted me. But I understood why it sounded mad.

We were scaling Blockdaemon from 10 to 187 engineers. We’d just closed a funding round. Everyone knows you’re supposed to lock down your best people during hypergrowth, not let them moonlight with the competition.

But here’s what I’d learned: certain engineers need challenging work more than they need loyalty speeches. If Blockdaemon couldn’t provide that challenge every single day, they’d leave anyway. Better to let them scratch that itch with Ethereum Foundation while bringing back knowledge that would benefit us for years.

New data from LeadDev shows 38% of engineering leaders report less motivated teams than a year ago. Only 21% describe their teams as healthy. 12% say they’re emotionally drained daily.

If you’re a founder or engineering leader reading those numbers and thinking “my team seems fine,” you might be six months behind reality.

I’m seeing this across my network right now. Good engineers going quiet in standups. High performers suddenly interviewing elsewhere. Teams that used to ship fast now dragging on simple tasks.

I prepared technical due diligence materials for Blockdaemon’s Series A, B, and C funding rounds. Three times I had to answer the question: how do we know your technology actually works at scale?

The answer that got us to $3.25 billion valuation wasn’t the one most CTOs think matters.

VCs don’t care about your velocity metrics. They don’t care about story points. They barely care about your deployment frequency.

What impressed them was simple: we were running infrastructure for Meta and JP Morgan across five continents, and it worked.

After Blockdaemon hit $3.25 billion valuation, I started talking to companies about permanent CTO roles.

None of them felt as challenging as what I’d just done. Scaling from 10 to 187 engineers. Building infrastructure that Meta and JP Morgan trusted. Deploying validators across five continents. Managing 96% retention through hypergrowth.

I could wait for another opportunity like Blockdaemon. But I’d probably wait forever.

So I went fractional instead.

Why fractional made sense

The decision wasn’t complicated. I’d spent years building hard-earned knowledge about scaling technical teams, infrastructure, and organisations. I could apply that knowledge to help multiple companies rather than waiting for one perfect opportunity.

When I told our CEO I wanted our best blockchain engineers to work part-time for Ethereum Foundation and Solana, he didn’t blink. Constantine trusted me. But I understood why it sounded mad.

We were scaling Blockdaemon from 10 to 187 engineers. We’d just closed a funding round. Everyone knows you’re supposed to lock down your best people during hypergrowth, not let them moonlight with the competition.

But here’s what I’d learned: certain engineers need challenging work more than they need loyalty speeches. If Blockdaemon couldn’t provide that challenge every single day, they’d leave anyway. Better to let them scratch that itch with Ethereum Foundation while bringing back knowledge that would benefit us for years.

I prepared technical due diligence materials for Blockdaemon’s Series A, B, and C funding rounds. Three times I had to answer the question: how do we know your technology actually works at scale?

The answer that got us to $3.25 billion valuation wasn’t the one most CTOs think matters.

VCs don’t care about your velocity metrics. They don’t care about story points. They barely care about your deployment frequency.

What impressed them was simple: we were running infrastructure for Meta and JP Morgan across five continents, and it worked.

I prepared technical due diligence materials for Blockdaemon’s Series A, B, and C funding rounds. Three times I had to answer the question: how do we know your technology actually works at scale?

The answer that got us to $3.25 billion valuation wasn’t the one most CTOs think matters.

VCs don’t care about your velocity metrics. They don’t care about story points. They barely care about your deployment frequency.

What impressed them was simple: we were running infrastructure for Meta and JP Morgan across five continents, and it worked.

I prepared technical due diligence materials for Blockdaemon’s Series A, B, and C funding rounds. Three times I had to answer the question: how do we know your technology actually works at scale?

The answer that got us to $3.25 billion valuation wasn’t the one most CTOs think matters.

VCs don’t care about your velocity metrics. They don’t care about story points. They barely care about your deployment frequency.

What impressed them was simple: we were running infrastructure for Meta and JP Morgan across five continents, and it worked.

After Blockdaemon hit $3.25 billion valuation, I started talking to companies about permanent CTO roles.

None of them felt as challenging as what I’d just done. Scaling from 10 to 187 engineers. Building infrastructure that Meta and JP Morgan trusted. Deploying validators across five continents. Managing 96% retention through hypergrowth.

I could wait for another opportunity like Blockdaemon. But I’d probably wait forever.

So I went fractional instead.

Why fractional made sense

The decision wasn’t complicated. I’d spent years building hard-earned knowledge about scaling technical teams, infrastructure, and organisations. I could apply that knowledge to help multiple companies rather than waiting for one perfect opportunity.

After Blockdaemon hit $3.25 billion valuation, I started talking to companies about permanent CTO roles.

None of them felt as challenging as what I’d just done. Scaling from 10 to 187 engineers. Building infrastructure that Meta and JP Morgan trusted. Deploying validators across five continents. Managing 96% retention through hypergrowth.

I could wait for another opportunity like Blockdaemon. But I’d probably wait forever.

So I went fractional instead.

Why fractional made sense

The decision wasn’t complicated. I’d spent years building hard-earned knowledge about scaling technical teams, infrastructure, and organisations. I could apply that knowledge to help multiple companies rather than waiting for one perfect opportunity.

When I told our CEO I wanted our best blockchain engineers to work part-time for Ethereum Foundation and Solana, he didn’t blink. Constantine trusted me. But I understood why it sounded mad.

We were scaling Blockdaemon from 10 to 187 engineers. We’d just closed a funding round. Everyone knows you’re supposed to lock down your best people during hypergrowth, not let them moonlight with the competition.

But here’s what I’d learned: certain engineers need challenging work more than they need loyalty speeches. If Blockdaemon couldn’t provide that challenge every single day, they’d leave anyway. Better to let them scratch that itch with Ethereum Foundation while bringing back knowledge that would benefit us for years.

When I told our CEO I wanted our best blockchain engineers to work part-time for Ethereum Foundation and Solana, he didn’t blink. Constantine trusted me. But I understood why it sounded mad.

We were scaling Blockdaemon from 10 to 187 engineers. We’d just closed a funding round. Everyone knows you’re supposed to lock down your best people during hypergrowth, not let them moonlight with the competition.

But here’s what I’d learned: certain engineers need challenging work more than they need loyalty speeches. If Blockdaemon couldn’t provide that challenge every single day, they’d leave anyway. Better to let them scratch that itch with Ethereum Foundation while bringing back knowledge that would benefit us for years.