Activity
Mon
Wed
Fri
Sun
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
What is this?
Less
More
28 contributions to AI Automation Society
When should an automation ask for help?
I've found the tricky part isn't always getting an automation to run. It's knowing when it should pause and let a person take a look. What's one situation where you've decided your automation should hand a decision back to a human?
3 likes • 1d
@Jason Elam One situation is when a caller needs a decision that depends on human judgment or information the system can’t verify. The automation can gather the details and explain why it’s handing the conversation over. Then the person stepping in has the context and the caller doesn’t have to start over.
What are you building with automation and hardware?
I’ve been thinking about automation that connects software to something people use in the real world. One idea I’m exploring is OrderBridge: phone orders would appear on a screen in the kitchen, and staff could confirm them there. I’m curious what others have created by combining automation with hardware—screens, tablets, sensors, phones, or other devices. What problem did it solve, and what possibilities did it open up that software alone couldn’t?
1 like • 3d
@Cainin Whisenant I agree that fast follow-up can matter more than adding hardware. I see the kitchen screen solving a different handoff: it gives staff a shared place to see and acknowledge an order. In your missed-call audits, do you also track how quickly the business calls back?
1 like • 3d
@Wasilis Paliakoudis That makes sense—a named person at each step gives the alert clear ownership and avoids the “someone else has it” problem. For the kitchen flow I’m exploring, that could mean notifying the shift lead first, then escalating if the order stays unacknowledged. How long do you usually wait before moving to the next person?
A green check is not the finish line
Finding out an automation has been running without delivering what anyone needed is frustrating. One lesson from systems engineering: a successful run and a useful result are two different checks. Now I define the outcome first, then ask what evidence would show it happened. For a follow-up workflow, that could mean checking whether the message reached the intended person, not only whether the send step completed. What's changed how you check your automations?
4 likes • 3d
@Jason Elam This is a distinction I’m applying as I think through OrderBridge. A successful workflow would only prove the order was sent. The real checks are whether it appeared on the kitchen screen, staff acknowledged it, and an unacknowledged order triggered another alert. I’m also considering whether the customer received confirmation. That helps define what a useful test should measure.
3 likes • 3d
@Geoff Bosley That’s a useful way to frame it. For the OrderBridge concept I’m exploring, both the kitchen staff and customer should benefit. I’d want to check more than whether an order reached the screen: did staff acknowledge it, was it accurate, and did the customer receive confirmation? Your point about human spot-checks is a good reminder that a system needs to catch mistakes, not just report activity. How do you keep those checks useful without adding too much work?
ANNOUNCING: A certification for AI builders who want bigger clients
Building with AI gets easier every month. Landing the bigger clients doesn't. Because the moment you pitch a growing company, the questions change. - Legal asks how you're handling data governance. - Finance asks what it costs to run at scale. - Procurement asks how they know it'll work reliably. Most builders don't have those answers. And that's where the deal dies. So we built a program to fix it. -> ANNOUNCING: A certification for AI builders who want bigger clients It's called the Certified AI Consultant program. Over 10 weeks, it teaches the 7 skills bigger companies actually pay for: scoping, pricing, governance, evals, AI economics, and more. Then you earn the credential through a graded portfolio and a live interview with my team. It's already working. Reidar, from our first cohort, closed a $75,000 contract with a global engineering company using what he learned. Cohort 2 starts November 2nd, and it's capped at 50 seats. Let me know if you have any questions. Talk soon, Nate
3 likes • 27d
How to join?
What Makes an AI Ready to Handle Customer Conversations?
I’m working through what it takes to make an AI system reliably handle routine customer conversations while recognizing when a person needs to step in. I’m especially interested in the practical lessons that don’t always show up in a demo: - How do you decide which tasks the AI can handle on its own? - What situations should trigger a handoff to a person? - How do you test whether it’s ready for real conversations, and what do you monitor after launch? If you’ve built something similar, what lesson saved you the most time or prevented the biggest mistake? And if you’re still figuring this out, what part is giving you the most trouble? Your answer may help others working through the same challenge.
1-10 of 28
Glen Lloyd
5
70 points to level up
@glen-lloyd-4630
On a mission to be the Automation King 👑. Building VoiceShield to help businesses grow. Open to collab in marketing + automation.

Active 6h ago
Joined Jan 7, 2026
Memphis,Tn
Powered by