Joel Beasley

CTO, Podcast Host, Author, Husband & Dad.

[Joel Beasley @moderncto_io]

The organizations winning at AI in HR aren't doing the coolest things.

They're doing the boring things well.

@SmartRecruiters backed the research because they kept seeing the same thing in the market: a lot of noise, not a lot of movement.

[Joel Beasley @moderncto_io]

HR's best risk management strategy is risk avoidance.

And that, according to researcher Kyle Lagunas, is why so many HR orgs are stuck.

@SmartRecruiters partnered with Kyle & Co to study what actually separates the organizations moving forward from those frozen in place.

[Joel Beasley @moderncto_io]

The easy problems are getting automated. All of them.

If your job is moving data from one box to another, good luck.

Jacques Klick from @wearesinch puts it plainly: the people who will succeed are the ones who run toward complexity while everyone else backs away from it.

[Joel Beasley @moderncto_io]

Strict AI regulations allow for more speed.

Kyle Lagunas and Allyn Bailey at @SmartRecruiters found something they didn't expect in their research: EU and UK companies, operating under the EU AI Act, were keeping pace with North America and in some areas accelerating past them.

[Joel Beasley @moderncto_io]

If your AI agent made a mistake right now, would you even know?

Not eventually. Right now.

Jacques Klick from @wearesinch says this is the question most companies can't answer, and it's exactly why so many agent deployments get rolled back after launch.

[Joel Beasley @moderncto_io]

That text from Verizon with a checkmark and their logo instead of a random phone number? That's RCS. And most brands have no idea it exists.

Jacques Klick from @wearesinch breaks down why this is the most underrated channel in marketing right now.

[Joel Beasley @moderncto_io]

62% of enterprise leaders already have AI agents running live in production.

That sounds like a win. Until you hear the next number: 74% of them had to roll it back.

Jacques Klick from @wearesinch found that the story on AI runs way deeper than what you've been told.

[Joel Beasley @moderncto_io]

"Can your robot do laundry?"

Seems like a simple question. The answer is... complicated.

The short version: the @1x_tech Neo will do your laundry. The longer version is worth hearing.

[Joel Beasley @moderncto_io]

Could you beat a robot in a fight if you had to?

Say your household robotic servant gets tired of cleaning up after you. If push comes to shove, can the average robot-owner actually take one of these down?

Dar Sleeper, VP at @1x_tech, says that won't be an issue.

[Joel Beasley @moderncto_io]

Everyone assumed robots would go to factories first. @1x_tech built theirs for your living room instead.

Not because it's cooler (though it is). Because it's smarter.

The home isn't a compromise. It's the fastest path to a robot that can do everything.

[Joel Beasley @moderncto_io]

Is this the worst take on AI music?

Judge for yourself.

Joel points out that @Trevorhinesley is one of the most AI-native CTOs he's ever had on the show — and then pulls up a website that says "100% human-made music."

[Joel Beasley @moderncto_io]

The gap between how humans think and how robots "think" just got a lot smaller.

And Dar from @1x_tech says they solved it faster than anyone expected.

[Joel Beasley @moderncto_io]

What if a sales rep could show a customer a working prototype before hanging up the call?

That's what @Trevorhinesley is building at @SoundstripeApp. They call it Backstage.

No engineers in the loop until the prototype is done. Just a nontechnical person prompting an agent.

[Joel Beasley @moderncto_io]

A migration project scoped at 1.5 years. 3 to 4 people. 18 months of work.

They ran an autonomous AI system for two weeks and got 50 to 60% of it done.

@Trevorhinesley walks through exactly how they built the harness around Claude Code to make it work.

[Joel Beasley @moderncto_io]

"We have a policy: there is no human-written code. Ever."

That's not a goal. That's the current reality at @SoundstripeApp

@TrevorHinesley, CTO, broke down what that actually looks like in practice and why it's working.

[Joel Beasley @moderncto_io]

"Give me this feature the competitor has" vs. "here's the problem our customers are experiencing."

Those two sentences produce completely different outcomes from your engineers.

The people who know what's now possible are the engineers.

[Joel Beasley @moderncto_io]

"We have more garbage being shipped than I ever remember."

That's Marty Cagan, one of the most respected names in product management, on what's actually happening as AI gets adopted across the industry.

He's not blaming AI. He's blaming the process people are using it on.

[Joel Beasley @moderncto_io]

Two kinds of building. Most teams only know one.

"Build to learn" is prototyping. Fast, cheap, disposable. You're testing assumptions, not shipping product.

"Build to earn" is production. Scalable, fault-tolerant, real.

[Joel Beasley @moderncto_io]

Germline editing gets labeled "playing God."

But what if preventing a child from being born with a debilitating disease is closer to feeding the hungry than it is to science fiction?

Hard questions deserve more than a reflexive no.

[Joel Beasley @moderncto_io]

For years, every CTO said the same thing: "Our bottleneck is how long it takes to build."

AI just proved them wrong.

The bottleneck was never the engineers. It was always the question of what to ask them to build in the first place.