@michaeldonohue.bsky.social
There is a lot of comparison between this AI wave and the dot-com era. I lived through part of the latter as an engineer, and the comparison feels a little off to me.
December 15, 2025 at 6:30 PM
There is a lot of comparison between this AI wave and the dot-com era. I lived through part of the latter as an engineer, and the comparison feels a little off to me.
I’ve sat through enough pitch meetings to know some people can really sell.
So when I see a “State of AI” report showing that 88% of companies are “using AI,” but only a fraction are benefiting, I can’t help but think that some of those buyers definitely met some of those sellers.
So when I see a “State of AI” report showing that 88% of companies are “using AI,” but only a fraction are benefiting, I can’t help but think that some of those buyers definitely met some of those sellers.
The State of AI: Global Survey 2025 | McKinsey
In this 2025 edition of the annual McKinsey Global Survey on AI, we look at the current trends that are driving real value from artificial intelligence.
www.mckinsey.com
December 12, 2025 at 5:27 PM
I’ve sat through enough pitch meetings to know some people can really sell.
So when I see a “State of AI” report showing that 88% of companies are “using AI,” but only a fraction are benefiting, I can’t help but think that some of those buyers definitely met some of those sellers.
So when I see a “State of AI” report showing that 88% of companies are “using AI,” but only a fraction are benefiting, I can’t help but think that some of those buyers definitely met some of those sellers.
I hosted a CMU founders + investors happy hour, and it ended up being one of the most energizing nights I’ve had in a while.
Huge thanks to Chenxi Wang for partnering with me on this one. She jumped in to co-host and split the food bill, which made the whole thing possible.
Huge thanks to Chenxi Wang for partnering with me on this one. She jumped in to co-host and split the food bill, which made the whole thing possible.
December 9, 2025 at 5:18 PM
I hosted a CMU founders + investors happy hour, and it ended up being one of the most energizing nights I’ve had in a while.
Huge thanks to Chenxi Wang for partnering with me on this one. She jumped in to co-host and split the food bill, which made the whole thing possible.
Huge thanks to Chenxi Wang for partnering with me on this one. She jumped in to co-host and split the food bill, which made the whole thing possible.
Low-code platforms and AI tools have enabled more teams to present products faster than ever.
At first, it seems like that would make it harder to know who understands what they’re building, like trying to find a needle in a bigger haystack. But the opposite is happening.
At first, it seems like that would make it harder to know who understands what they’re building, like trying to find a needle in a bigger haystack. But the opposite is happening.
December 1, 2025 at 6:16 PM
Low-code platforms and AI tools have enabled more teams to present products faster than ever.
At first, it seems like that would make it harder to know who understands what they’re building, like trying to find a needle in a bigger haystack. But the opposite is happening.
At first, it seems like that would make it harder to know who understands what they’re building, like trying to find a needle in a bigger haystack. But the opposite is happening.
Most founders promise monthly investor updates. Almost none keep that pace.
That’s fine.
The goal was never to impress with frequency. Just communicate consistently.
For most situations, quarterly is plenty. Early-stage progress takes time. Shifts don't happen every 30 days.
That’s fine.
The goal was never to impress with frequency. Just communicate consistently.
For most situations, quarterly is plenty. Early-stage progress takes time. Shifts don't happen every 30 days.
November 25, 2025 at 6:15 PM
Most founders promise monthly investor updates. Almost none keep that pace.
That’s fine.
The goal was never to impress with frequency. Just communicate consistently.
For most situations, quarterly is plenty. Early-stage progress takes time. Shifts don't happen every 30 days.
That’s fine.
The goal was never to impress with frequency. Just communicate consistently.
For most situations, quarterly is plenty. Early-stage progress takes time. Shifts don't happen every 30 days.
Some of the best outcomes I’ve seen came from teams that didn’t get it right the first time.
A few years ago, I backed a team that couldn’t seem to find its footing. In their first six months, they changed their product, their customer, and even their pricing model three times.
A few years ago, I backed a team that couldn’t seem to find its footing. In their first six months, they changed their product, their customer, and even their pricing model three times.
November 13, 2025 at 6:15 PM
Some of the best outcomes I’ve seen came from teams that didn’t get it right the first time.
A few years ago, I backed a team that couldn’t seem to find its footing. In their first six months, they changed their product, their customer, and even their pricing model three times.
A few years ago, I backed a team that couldn’t seem to find its footing. In their first six months, they changed their product, their customer, and even their pricing model three times.
People assume WhatsApp scaled by burning VC money. In reality, we were cash-flow positive from Day 1.
I don’t share this to brag. I share it because the startup myth that “you can’t grow without burning” is still everywhere. And I’ve seen it sink good companies.
I don’t share this to brag. I share it because the startup myth that “you can’t grow without burning” is still everywhere. And I’ve seen it sink good companies.
November 12, 2025 at 6:32 PM
People assume WhatsApp scaled by burning VC money. In reality, we were cash-flow positive from Day 1.
I don’t share this to brag. I share it because the startup myth that “you can’t grow without burning” is still everywhere. And I’ve seen it sink good companies.
I don’t share this to brag. I share it because the startup myth that “you can’t grow without burning” is still everywhere. And I’ve seen it sink good companies.
Data centers are where the AI magic happens. But operations inside them are still surprisingly analog.
If a server fails today, someone walks an aisle with a clipboard, locates the correct rack, swaps the part and continues moving. It works, but it doesn’t scale with the demand.
If a server fails today, someone walks an aisle with a clipboard, locates the correct rack, swaps the part and continues moving. It works, but it doesn’t scale with the demand.
November 9, 2025 at 10:01 PM
Data centers are where the AI magic happens. But operations inside them are still surprisingly analog.
If a server fails today, someone walks an aisle with a clipboard, locates the correct rack, swaps the part and continues moving. It works, but it doesn’t scale with the demand.
If a server fails today, someone walks an aisle with a clipboard, locates the correct rack, swaps the part and continues moving. It works, but it doesn’t scale with the demand.
When Apple rolled out its “do not track” change, two of the CPG startups I’d invested in went from strong traction to stalled growth almost overnight.
Their Facebook ads stopped converting. And without another engine for acquisition, the business models couldn’t keep up.
Their Facebook ads stopped converting. And without another engine for acquisition, the business models couldn’t keep up.
October 30, 2025 at 4:47 PM
When Apple rolled out its “do not track” change, two of the CPG startups I’d invested in went from strong traction to stalled growth almost overnight.
Their Facebook ads stopped converting. And without another engine for acquisition, the business models couldn’t keep up.
Their Facebook ads stopped converting. And without another engine for acquisition, the business models couldn’t keep up.
Back when we were developing WhatsApp, one of the biggest unlocks came from something few people still think about: BlackBerry. Yes, that BlackBerry.
BBM (BlackBerry Messenger) was the standard, and it had already taught users what “always-on messaging” felt like.
BBM (BlackBerry Messenger) was the standard, and it had already taught users what “always-on messaging” felt like.
October 23, 2025 at 4:30 PM
Back when we were developing WhatsApp, one of the biggest unlocks came from something few people still think about: BlackBerry. Yes, that BlackBerry.
BBM (BlackBerry Messenger) was the standard, and it had already taught users what “always-on messaging” felt like.
BBM (BlackBerry Messenger) was the standard, and it had already taught users what “always-on messaging” felt like.
Referrals are a kind of currency in early-stage investing.
The best introductions I've gotten come from people I’ve worked with before. Someone who’s seen how a founder operates day to day carries weight a cold call or email never could.
The best introductions I've gotten come from people I’ve worked with before. Someone who’s seen how a founder operates day to day carries weight a cold call or email never could.
October 17, 2025 at 5:22 PM
Referrals are a kind of currency in early-stage investing.
The best introductions I've gotten come from people I’ve worked with before. Someone who’s seen how a founder operates day to day carries weight a cold call or email never could.
The best introductions I've gotten come from people I’ve worked with before. Someone who’s seen how a founder operates day to day carries weight a cold call or email never could.
Startup Week at @cmu.edu was a blast. Old friends, new founders, and good conversations.
We ended up at the Mexican restaurant Duolingo’s CEO insisted on putting in because Pittsburgh “didn’t have a good one.” That’s how problems get solved... and startups get built.
We ended up at the Mexican restaurant Duolingo’s CEO insisted on putting in because Pittsburgh “didn’t have a good one.” That’s how problems get solved... and startups get built.
October 7, 2025 at 7:28 PM
Startup Week at @cmu.edu was a blast. Old friends, new founders, and good conversations.
We ended up at the Mexican restaurant Duolingo’s CEO insisted on putting in because Pittsburgh “didn’t have a good one.” That’s how problems get solved... and startups get built.
We ended up at the Mexican restaurant Duolingo’s CEO insisted on putting in because Pittsburgh “didn’t have a good one.” That’s how problems get solved... and startups get built.
Everyone can sell the vision. Fewer bring the numbers to back it up.
Even early revenue, even if small, tells a story about who’s willing to pay, why the problem matters, and how the model might scale.
Sharing those figures shows discipline and judgment. That’s how founders earn trust.
Even early revenue, even if small, tells a story about who’s willing to pay, why the problem matters, and how the model might scale.
Sharing those figures shows discipline and judgment. That’s how founders earn trust.
September 25, 2025 at 5:16 PM
Everyone can sell the vision. Fewer bring the numbers to back it up.
Even early revenue, even if small, tells a story about who’s willing to pay, why the problem matters, and how the model might scale.
Sharing those figures shows discipline and judgment. That’s how founders earn trust.
Even early revenue, even if small, tells a story about who’s willing to pay, why the problem matters, and how the model might scale.
Sharing those figures shows discipline and judgment. That’s how founders earn trust.
I never interviewed at WhatsApp. A founder I’d met years earlier vouched for me, and that trust opened the door.
I was privileged, but that experience taught me something I’ve seen repeatedly in startups. Referrals are the currency of early-stage ecosystems, not just hiring shortcuts.
I was privileged, but that experience taught me something I’ve seen repeatedly in startups. Referrals are the currency of early-stage ecosystems, not just hiring shortcuts.
September 23, 2025 at 4:23 PM
I never interviewed at WhatsApp. A founder I’d met years earlier vouched for me, and that trust opened the door.
I was privileged, but that experience taught me something I’ve seen repeatedly in startups. Referrals are the currency of early-stage ecosystems, not just hiring shortcuts.
I was privileged, but that experience taught me something I’ve seen repeatedly in startups. Referrals are the currency of early-stage ecosystems, not just hiring shortcuts.
One of the most rewarding founder journeys I’ve been a part of began with a pitch I initially turned down.
A CMU grad pitched astronaut helmet displays. Interesting tech, but not a market I’d invest in.
A CMU grad pitched astronaut helmet displays. Interesting tech, but not a market I’d invest in.
September 16, 2025 at 5:23 PM
One of the most rewarding founder journeys I’ve been a part of began with a pitch I initially turned down.
A CMU grad pitched astronaut helmet displays. Interesting tech, but not a market I’d invest in.
A CMU grad pitched astronaut helmet displays. Interesting tech, but not a market I’d invest in.
One mistake I see early #Founders make is cluttering the #CapTable with dozens of small checks.
It feels like momentum, but it can add friction in future rounds.
Clean cap tables make life easier for everyone. It means more flexibility and fewer headaches when it matters most.
It feels like momentum, but it can add friction in future rounds.
Clean cap tables make life easier for everyone. It means more flexibility and fewer headaches when it matters most.
August 28, 2025 at 3:53 PM
One mistake I see early #Founders make is cluttering the #CapTable with dozens of small checks.
It feels like momentum, but it can add friction in future rounds.
Clean cap tables make life easier for everyone. It means more flexibility and fewer headaches when it matters most.
It feels like momentum, but it can add friction in future rounds.
Clean cap tables make life easier for everyone. It means more flexibility and fewer headaches when it matters most.
Just because you own 60% of your startup doesn’t mean you’re in control.
Control = voting rights, not just equity.
Super-voting shares help, but you have to ask for them early. Your lawyer won’t assume.
Small detail. Big leverage.
Control = voting rights, not just equity.
Super-voting shares help, but you have to ask for them early. Your lawyer won’t assume.
Small detail. Big leverage.
August 20, 2025 at 4:21 PM
Just because you own 60% of your startup doesn’t mean you’re in control.
Control = voting rights, not just equity.
Super-voting shares help, but you have to ask for them early. Your lawyer won’t assume.
Small detail. Big leverage.
Control = voting rights, not just equity.
Super-voting shares help, but you have to ask for them early. Your lawyer won’t assume.
Small detail. Big leverage.
Friends + family are great for support, but not for feedback.
If they’re your first “customers,” you’re not testing. You’re getting affirmation.
Real insight starts when someone can say no. #startups #earlystage #feedbackloop
If they’re your first “customers,” you’re not testing. You’re getting affirmation.
Real insight starts when someone can say no. #startups #earlystage #feedbackloop
August 11, 2025 at 2:27 PM
Friends + family are great for support, but not for feedback.
If they’re your first “customers,” you’re not testing. You’re getting affirmation.
Real insight starts when someone can say no. #startups #earlystage #feedbackloop
If they’re your first “customers,” you’re not testing. You’re getting affirmation.
Real insight starts when someone can say no. #startups #earlystage #feedbackloop
If you’re still working full-time and say, “I’ll quit once I raise,” that’s a red flag for investors.
Startups are hard. If you’re not ready to take the first risk, don’t expect others to.
Believe enough to go all-in, or wait until you do.
#Startups #Fundraising
Startups are hard. If you’re not ready to take the first risk, don’t expect others to.
Believe enough to go all-in, or wait until you do.
#Startups #Fundraising
August 5, 2025 at 3:30 PM
If you’re still working full-time and say, “I’ll quit once I raise,” that’s a red flag for investors.
Startups are hard. If you’re not ready to take the first risk, don’t expect others to.
Believe enough to go all-in, or wait until you do.
#Startups #Fundraising
Startups are hard. If you’re not ready to take the first risk, don’t expect others to.
Believe enough to go all-in, or wait until you do.
#Startups #Fundraising
At top #engineering schools, the culture is more about getting hired than building something new.
I saw it at Carnegie Mellon. Every fall, the job fair takes over campus. #BigTech internships are the badge of honor. Starting a company? That’s more of an afterthought.
I saw it at Carnegie Mellon. Every fall, the job fair takes over campus. #BigTech internships are the badge of honor. Starting a company? That’s more of an afterthought.
July 18, 2025 at 4:41 PM
At top #engineering schools, the culture is more about getting hired than building something new.
I saw it at Carnegie Mellon. Every fall, the job fair takes over campus. #BigTech internships are the badge of honor. Starting a company? That’s more of an afterthought.
I saw it at Carnegie Mellon. Every fall, the job fair takes over campus. #BigTech internships are the badge of honor. Starting a company? That’s more of an afterthought.
#AI might kill more #VC firms than startups.
#ZombieFunds are up 50% since 2021.
Some say AI is to blame - founders don’t need as much capital.
But it’s not less capital that matters.
It’s smarter capital.
If your only value is a check, good luck.
#ZombieFunds are up 50% since 2021.
Some say AI is to blame - founders don’t need as much capital.
But it’s not less capital that matters.
It’s smarter capital.
If your only value is a check, good luck.
AI-built startups could fuel increase in venture capital ‘zombie’ funds
The number of zombie VC funds has jumped by 50% since the end of 2021.
fortune.com
July 9, 2025 at 4:31 PM
#AI might kill more #VC firms than startups.
#ZombieFunds are up 50% since 2021.
Some say AI is to blame - founders don’t need as much capital.
But it’s not less capital that matters.
It’s smarter capital.
If your only value is a check, good luck.
#ZombieFunds are up 50% since 2021.
Some say AI is to blame - founders don’t need as much capital.
But it’s not less capital that matters.
It’s smarter capital.
If your only value is a check, good luck.
If the person who can say yes isn’t in the room, don’t have the meeting.
You’re wasting the #pitch or handing off your story for someone else to retell.
It sounds obvious, but the urge to #demo is strong.
Wait for the right moment, with the right person in the room.
You’re wasting the #pitch or handing off your story for someone else to retell.
It sounds obvious, but the urge to #demo is strong.
Wait for the right moment, with the right person in the room.
June 25, 2025 at 2:23 PM
If the person who can say yes isn’t in the room, don’t have the meeting.
You’re wasting the #pitch or handing off your story for someone else to retell.
It sounds obvious, but the urge to #demo is strong.
Wait for the right moment, with the right person in the room.
You’re wasting the #pitch or handing off your story for someone else to retell.
It sounds obvious, but the urge to #demo is strong.
Wait for the right moment, with the right person in the room.
Every startup pitch has that one slide with logos.
You know the one:
– “We’re already talking to these Fortune 500s”
– “These are our pipeline targets”
– “We’ve had great initial conversations”
You know the one:
– “We’re already talking to these Fortune 500s”
– “These are our pipeline targets”
– “We’ve had great initial conversations”
June 20, 2025 at 5:15 PM
Every startup pitch has that one slide with logos.
You know the one:
– “We’re already talking to these Fortune 500s”
– “These are our pipeline targets”
– “We’ve had great initial conversations”
You know the one:
– “We’re already talking to these Fortune 500s”
– “These are our pipeline targets”
– “We’ve had great initial conversations”
Creator jobs in the U.S. have grown 7.5x since 2020, per Axios. We’ve been in the wave for a bit, and while growth like that doesn’t guarantee returns, it’s a clear signal the space is real.
Models are maturing, talent’s still pouring in, and still a lot of room to build.
#CreatorEconomy #Startups
Models are maturing, talent’s still pouring in, and still a lot of room to build.
#CreatorEconomy #Startups
Exclusive: Digital creator jobs jump from 200k to 1.5 million since pandemic
New tools have lowered the bar to entry to make and monetize content.
www.axios.com
June 4, 2025 at 3:43 PM
Creator jobs in the U.S. have grown 7.5x since 2020, per Axios. We’ve been in the wave for a bit, and while growth like that doesn’t guarantee returns, it’s a clear signal the space is real.
Models are maturing, talent’s still pouring in, and still a lot of room to build.
#CreatorEconomy #Startups
Models are maturing, talent’s still pouring in, and still a lot of room to build.
#CreatorEconomy #Startups
#AI cooled to ~30% of global venture funding in April, per Crunchbase News.
Still more share than #crypto or #SaaS at their peaks.
You’d have to go back to the mobile boom for a #tech trend this dominant. Even then, capital wasn’t this concentrated.
Still more share than #crypto or #SaaS at their peaks.
You’d have to go back to the mobile boom for a #tech trend this dominant. Even then, capital wasn’t this concentrated.
Global Venture Funding Slowed In April, Despite Strong AI Showing
Global venture funding totaled $23 billion in April 2025, flat year over year and down significantly month over month from $68 billion invested in March, Crunchbase data shows.
news.crunchbase.com
May 29, 2025 at 3:12 PM
#AI cooled to ~30% of global venture funding in April, per Crunchbase News.
Still more share than #crypto or #SaaS at their peaks.
You’d have to go back to the mobile boom for a #tech trend this dominant. Even then, capital wasn’t this concentrated.
Still more share than #crypto or #SaaS at their peaks.
You’d have to go back to the mobile boom for a #tech trend this dominant. Even then, capital wasn’t this concentrated.