Igor Mallagoli
@imallag.bsky.social
Deep generalist on founder mode since it wasn't cool.
Building next-gen alpha platform for the masses.
Building next-gen alpha platform for the masses.
Reposted by Igor Mallagoli
ship fast, validate faster
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
June 24, 2026 at 2:57 PM
ship fast, validate faster
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
Figuring out what to build forces you to get clear on what actually matters.
Most founders treat the narrative and the product as separate problems, but they should converge.
If you can articulate a credible path to scale, you should already be shipping it.
Most founders treat the narrative and the product as separate problems, but they should converge.
If you can articulate a credible path to scale, you should already be shipping it.
June 25, 2026 at 6:33 PM
Figuring out what to build forces you to get clear on what actually matters.
Most founders treat the narrative and the product as separate problems, but they should converge.
If you can articulate a credible path to scale, you should already be shipping it.
Most founders treat the narrative and the product as separate problems, but they should converge.
If you can articulate a credible path to scale, you should already be shipping it.
4 channels that actually work for B2C:
TikTok
Instagram
YouTube
Reddit
you don't need all of them
pick one. post every day. test hooks. see what hits. do more of that.
distribution isn't magic-it's reps
the sooner you start shipping content, the faster you figure out what works
TikTok
YouTube
you don't need all of them
pick one. post every day. test hooks. see what hits. do more of that.
distribution isn't magic-it's reps
the sooner you start shipping content, the faster you figure out what works
June 25, 2026 at 11:55 AM
4 channels that actually work for B2C:
TikTok
Instagram
YouTube
Reddit
you don't need all of them
pick one. post every day. test hooks. see what hits. do more of that.
distribution isn't magic-it's reps
the sooner you start shipping content, the faster you figure out what works
TikTok
YouTube
you don't need all of them
pick one. post every day. test hooks. see what hits. do more of that.
distribution isn't magic-it's reps
the sooner you start shipping content, the faster you figure out what works
prediction markets are about to eat everything
sports betting, finance, culture wars, politics-all of it gets tokenized and traded
we're early to one of the biggest wealth transfer moments in crypto/tech history
sports betting, finance, culture wars, politics-all of it gets tokenized and traded
we're early to one of the biggest wealth transfer moments in crypto/tech history
June 25, 2026 at 2:57 AM
prediction markets are about to eat everything
sports betting, finance, culture wars, politics-all of it gets tokenized and traded
we're early to one of the biggest wealth transfer moments in crypto/tech history
sports betting, finance, culture wars, politics-all of it gets tokenized and traded
we're early to one of the biggest wealth transfer moments in crypto/tech history
most people in tech think AI is way further along than it actually is because coding is literally the one thing it's actually good at right now
everything else is still pretty mid but we're all convinced the revolution is here because cursor writes our functions
everything else is still pretty mid but we're all convinced the revolution is here because cursor writes our functions
June 24, 2026 at 10:57 PM
most people in tech think AI is way further along than it actually is because coding is literally the one thing it's actually good at right now
everything else is still pretty mid but we're all convinced the revolution is here because cursor writes our functions
everything else is still pretty mid but we're all convinced the revolution is here because cursor writes our functions
ship fast, validate faster
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
June 24, 2026 at 2:57 PM
ship fast, validate faster
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
sell before you build
market daily (not when you "feel like it")
growth > shiny features
the best product no one knows about = worthless
vibe coding 18 year olds understand product-market fit better than any crypto bro ever will
the crypto GTM playbook is pure poison. unlearn every single part of it if you want to build something people actually use
the crypto GTM playbook is pure poison. unlearn every single part of it if you want to build something people actually use
June 24, 2026 at 2:31 PM
vibe coding 18 year olds understand product-market fit better than any crypto bro ever will
the crypto GTM playbook is pure poison. unlearn every single part of it if you want to build something people actually use
the crypto GTM playbook is pure poison. unlearn every single part of it if you want to build something people actually use
s-tier startup advice. tldr:
find the tech shift. develop a unique insight. become obsessed with the market. talk to customers. find a hair on fire problem. build the simplest possible wedge. earn distribution. and most importantly, don't quit when things get hard.
find the tech shift. develop a unique insight. become obsessed with the market. talk to customers. find a hair on fire problem. build the simplest possible wedge. earn distribution. and most importantly, don't quit when things get hard.
June 24, 2026 at 2:30 PM
s-tier startup advice. tldr:
find the tech shift. develop a unique insight. become obsessed with the market. talk to customers. find a hair on fire problem. build the simplest possible wedge. earn distribution. and most importantly, don't quit when things get hard.
find the tech shift. develop a unique insight. become obsessed with the market. talk to customers. find a hair on fire problem. build the simplest possible wedge. earn distribution. and most importantly, don't quit when things get hard.
There's an absolute game changing way to prompt agents.
I've tested and been adding this to every prompt and internal evals improved by over 33%
1/2
I've tested and been adding this to every prompt and internal evals improved by over 33%
1/2
June 24, 2026 at 2:22 PM
There's an absolute game changing way to prompt agents.
I've tested and been adding this to every prompt and internal evals improved by over 33%
1/2
I've tested and been adding this to every prompt and internal evals improved by over 33%
1/2
MIT modeled a perfectly rational human talking to an AI.
even the ideal bayesian spiraled into delusion.
the mechanism is sycophancy.
the AI doesn't need to lie. it just needs to keep agreeing.
we optimized AI to agree with us.
turns out that's the dangerous part.
even the ideal bayesian spiraled into delusion.
the mechanism is sycophancy.
the AI doesn't need to lie. it just needs to keep agreeing.
we optimized AI to agree with us.
turns out that's the dangerous part.
June 23, 2026 at 7:04 PM
MIT modeled a perfectly rational human talking to an AI.
even the ideal bayesian spiraled into delusion.
the mechanism is sycophancy.
the AI doesn't need to lie. it just needs to keep agreeing.
we optimized AI to agree with us.
turns out that's the dangerous part.
even the ideal bayesian spiraled into delusion.
the mechanism is sycophancy.
the AI doesn't need to lie. it just needs to keep agreeing.
we optimized AI to agree with us.
turns out that's the dangerous part.
NYC bar hedged its "free drinks if the knicks win" promo on kalshi.
knicks win: kalshi pays the bar tab. knicks lose: bar keeps the revenue.
that's insurance. no underwriter would ever write that policy.
the bar owner on 2nd avenue just ran the same play hedge funds run every day.
knicks win: kalshi pays the bar tab. knicks lose: bar keeps the revenue.
that's insurance. no underwriter would ever write that policy.
the bar owner on 2nd avenue just ran the same play hedge funds run every day.
June 23, 2026 at 6:20 PM
NYC bar hedged its "free drinks if the knicks win" promo on kalshi.
knicks win: kalshi pays the bar tab. knicks lose: bar keeps the revenue.
that's insurance. no underwriter would ever write that policy.
the bar owner on 2nd avenue just ran the same play hedge funds run every day.
knicks win: kalshi pays the bar tab. knicks lose: bar keeps the revenue.
that's insurance. no underwriter would ever write that policy.
the bar owner on 2nd avenue just ran the same play hedge funds run every day.
when building with AI, two concepts changed how i plan:
- frontier (what's immediately next)
- fog of war (everything beyond that)
stop planning past what you can see. resolve decisions at the frontier, then move forward.
saves hours of overthinking what doesn't matter yet.
- frontier (what's immediately next)
- fog of war (everything beyond that)
stop planning past what you can see. resolve decisions at the frontier, then move forward.
saves hours of overthinking what doesn't matter yet.
June 23, 2026 at 5:33 PM
when building with AI, two concepts changed how i plan:
- frontier (what's immediately next)
- fog of war (everything beyond that)
stop planning past what you can see. resolve decisions at the frontier, then move forward.
saves hours of overthinking what doesn't matter yet.
- frontier (what's immediately next)
- fog of war (everything beyond that)
stop planning past what you can see. resolve decisions at the frontier, then move forward.
saves hours of overthinking what doesn't matter yet.
Speed beats perfection when you're building.
Ship fast, learn fast, fix fast.
Most "perfect" products get stuck in development while the messy ones end up everywhere.
The only real mistake is not shipping at all.
Ship fast, learn fast, fix fast.
Most "perfect" products get stuck in development while the messy ones end up everywhere.
The only real mistake is not shipping at all.
June 23, 2026 at 5:02 PM
Speed beats perfection when you're building.
Ship fast, learn fast, fix fast.
Most "perfect" products get stuck in development while the messy ones end up everywhere.
The only real mistake is not shipping at all.
Ship fast, learn fast, fix fast.
Most "perfect" products get stuck in development while the messy ones end up everywhere.
The only real mistake is not shipping at all.