Paul C. Chambers

The hardest part of the agent era isn’t the technology. It’s deciding which decisions you’re comfortable handing off. Inbox zero? Easy call. Client communication? Harder. Strategic judgment? Not yet. The operators getting this right are drawing those lines deliberately — not letting the tools make the call for them.

@coreyganim Operators who learn to build with AI stop depending on vendors to solve their problems. That's the real shift — from consumer to builder. Watching this closely.

been building this for months. finally ready to show it. Build With AI opens Tuesday. here's a look inside the classroom 👇

@jspeiser Same pattern ran through subscription commerce circa 2020. Operators who panicked productively rebuilt their workflows; the ones who waited still haven't caught up. "Yes, and" always wins.

A large dev shop founder in Hampton posted: "AI is affecting development more than anything. We have to stay paranoid about evolving our business." Within hours, 10+ agency founders in Hampton signed up for a self-org'd mastermind group. That panic is productive. The dev shops that survive won't be the ones pretending AI isn't coming for their margins. They'll be the ones who said "yes, and" rebuilt around it.

@steipete Keep-alive/daemon reliability is the one that blocked production use for us. These updates aren't glamorous — they're the ones that actually let you ship.

MCPorter (MCP->CLI)🧳0.8.0 is out. - stronger OAuth handling for servers - valid JSON output on fallback paths - better mcporter call behavior and error handling - generated CLIs handle object-valued args better - keep-alive/daemon reliability github.com/steipete/mcpor…

@minchoi Operators running agent workflows care less about who wins the compute war than about API stability and costs. Infrastructure bets matter more than model bets right now.

Google is getting deeper into Anthropic. FT says it plans to help finance a giant Texas data center. The real AI war is now compute + power👇

@alliekmiller The no-laptop hack day is a real forcing function. You find out fast which AI workflows are actually embedded vs. which ones are just a desk habit dressed up as automation.

I don’t think you all understand how powerful Claude Code is from a phone. Telegram, iMessage, remote control, mobile command, dispatch, whatever you have to do. Just get this thing on your phone. My team is going to hold a no-laptop hack day soon, where everyone is forced to leave their desk and just be on a walk or at a restaurant or in a park with their phone (and dictation, I’m sure). Whoever is most productive wins.

😑🙄

We crossed a line somewhere. AI stopped being a research tool and started being a team member. Not metaphorically. It’s filing ops, sending emails, managing calendars. We’re in a different category now.

@yojimmykim Decision fatigue is one of the most underrated churn drivers in subscription. Every unnecessary choice before the billing date is friction. Reduce the decisions, reduce the churn.

Here’s something subtle that kills conversions and retention: You make people decide… then decide again. Example: 1. Decide to click 2. Decide to buy 3. Decide which version 4. Decide when to use it 5. Decide when to reorder That’s exhausting. Strong brands collapse decisions. They imply, “This is just what you do next” Operationally, this looks like: - Default bundles instead of optional add-ons - “Most people reorder here” instead of “Would you like to reorder?” - One recommended path, not five choices People hate deciding. Make it easier.

@TaylorHoliday More interesting question: are you tracking token ROI? Most operators still treat AI spend as a fixed line item, not a variable tied to output. Until you measure that, any percentage is a guess.

What percent of salary should token utilization be per employee?

@archit_hacks "Works like you" is where most of these fail. Not mimicking output — it's knowing context. What you'd never say. Who to escalate to. Why a 6pm email reads differently than a 9am one.

Introducing Stamp: The AI Secretary that thinks, writes, and works like you.

@RampLabs Not AI that writes copy — AI that handles approvals, receipts, and vendor payments. Finance workflows are where agents actually earn their keep. This is the category operators needed.

Today, we're releasing Ramp CLI to let agents manage your company's finances. 50+ tools across cards, bills, expenses, travel, and approvals. Fewer tokens than MCP, and comes with pre-built skills like receipt compliance and agentic purchasing.

@gregisenberg Distribution has always been the moat — AI just made it obvious. In subscription, the brands winning aren't the ones with the best product. They're the ones who keep showing up after the purchase.

now that we all can build anything with ai we're going to all have to figure out distribution the wealthiest people will be marketers over the next 10 years

AI agents aren't a content play — that's table stakes. The subscription operators pulling ahead are using them for decisions: who gets a retention offer today, who gets a pause option, when to trigger the win-back. The leverage is in the judgment calls.

Lots of companies are now building primitives for an economy where AI agents are the primary users instead of humans. They're betting on an economy of AI coworkers. 1. AgentMail (@agentmail): so agents can have email accounts 2. AgentPhone (@tryagentphone): so agents can have phone numbers 3. Kapso (@andresmatte): so agents can have WhatsApp phone numbers 4. Daytona (@daytonaio) / E2B (@e2b): so agents can have their own computers 5. Browserbase (@browserbase) / Browser Use (@browser_use) / Hyperbrowser (@hyperbrowser): so agents can use web browsers 6. Firecrawl (@firecrawl): so agents can crawl the web without a browser 7. Mem0 (@mem0ai): so agents can remember things 8. Kite (@GoKiteAI) / Sponge (@PayspongeLabs) : so agents can pay for things. 9. Composio (@composio): so agents can use your SaaS tools 10. Orthogonal (@orthogonal_sh) so agents can access APIs easily 11. ElevenLabs (@ElevenLabs) / Vapi (@Vapi_AI) so agents can have a voice 12. Sixtyfour (@sixtyfourai) so agents can search for people and companies. 13. Exa (@ExaAILabs): so agents can search the web (Google doesn’t work for agents) If you stitch all of these together, you get a digital coworker that looks more human than AI.

@austin_rief 😂

Today I'm excited to introduce Hark, a new artificial intelligence lab building the most advanced, personal intelligence in the world We've been in stealth for 8 months, assembling one of the greatest AI and hardware teams on the planet I want to explain why I started Hark and what we're focused on I've spent the last 3 years working on the hardest AI challenge imaginable: giving AI a humanoid body. On the digital side, I've been using all the existing LLM chatbots - and I have to say, they feel incredibly dumb to me AGI, in the limit, should feel like a sci-fi movie. It should be able to listen and talk. It should have persistent memory and be highly personalized. It should see and touch the world. But we're far from this today We are crafting a new interface to AGI. Intelligence that lets you offload your mental workload into a system that begins to think like you and sometimes ahead of you We started Hark with one goal: build the world's most advanced personal intelligence - paired with next-generation hardware designed to serve as a universal interface between humans and machines hark.com

Costco CEO Ron Vachris just ate the $1.50 hot dog on camera and swore: “The price will not change as long as I’m around.” 🌭 Unchanged since 1985. Peak membership strategy — tiny perk, massive loyalty & renewals.

@JesseTinsley @MainStreet 😳🤩

@MainStreet completed 13 acquisitions in 3 years from 2023 to 2025. We’re targeting 20 acquisitions in 2026. Median acquisition price 2025: $22 million Median acquisition offer 2026: $91 million If you’re a banker, founder, VC, employee at an early stage startup and your or your company are exploring g potential exits in 2026 or 2027 we should talk.

@mcuban @IrenaCronin Agent Authorization sounds like a business.

In the near future. the marginal cost to create and run an agent will be minimal, so unlimited numbers of agents will compete in what appears to be an absolutely efficient market. However, there will be too many competitors. Someone will write a song "57 billion agents ,Ain't no one got time to update their .MD files to decide which is best , so I worked with my friend's company. I trust him more " The real currency will be trust. The paradox of choice will overwhelm everyone. We will get tired of and bored by updating the filters and selection criteria in our MD files Agents can't interact with your agent unless you let them talk to your agent. It will be as if you have to update your own spam filter. Yes it can be automated to a degree. But it's your responsibility to determine the exceptions. Kind of like you do know when the phone rings. That will be a pain in the ass