Zapier

I think one of the things that might prevent people from building with AI is feeling like they need big, high impact, flashy things to show for it. I am not that skilled, I haven't built anything that's gonna change the world. But I have built many agents that take care of small annoying tasks for me. One that I got to test out today is to help me with putting up OOO messages! There's about a 30% likelihood that I remember to set up my OOO responder while on vacation or to put up an OOO status on Slack. It's not a huge issue, but can mean that people are wondering why I'm not being responsive. So I built an agent that checks my calendar, and when it sees I have an OOO coming up: 1. Sends me an email reminding me to set my OOO on gmail. 2. Automatically updates my Slack status. This only took a few minutes to build, it's not going to save me a crazy amount of time, but it'll save me a little bit of mental labor, and make me appear a bit more organized. I built this agent using Zapier copilot which is helpful because you don't have to build your prompt and it does any troubleshooting for you. And this week, I got to test this out and just as expected, I received an email yesterday reminding me to set my auto-responder, and my status was automatically updated! These minor use cases are a great way to build your skill in low-risk ways - it's not going to negatively impact your work or anyone else if it doesn't work properly, but along the way, you'll learn more around how to trigger, build actions, what limitations you might run into, which apps you'll need to pull in to achieve different outcomes, how the agents "think" and troubleshoot, etc. For example, in building this, I learned: - AI agents CAN'T access your email settings (because my original goal was for it to set my auto-responded automatically) so the email reminder was a workaround. - Setting my status didn't automatically result in my agent updating me to away, I'll need to tweak those instructions. - I was hoping my status would show a return date, I'll need to clarify my instructions further to make that happen. So anyway, if you've got some free time over the holidays, pick some minor annoyance in your life, and see if you can build an agent that can handle all of that for you! P.S. I instructed my agent to consider holidays when selecting an emoji and I think it nailed it!!

If you do something repeatedly … automate it. (steal this system ↓) Step 1: Audit your time Write down every task you do in a week. Here’s a free template I created to help you — https://lnkd.in/e6ZTercE Step 2: Decide how to handle each task Delegate it → if it needs human judgment Automate it → if it follows rules Own it → if it's core to you (don’t get stuck here because you think everything needs you to be done right … it doesn’t) Step 3: Start small Pick the annoying stuff that steals focus. Those 2 minute tasks that interrupt deep work. Here’s one I made using Zapier ↓ Every hour, it scans Slack for messages with "due," "by," or "end of day." Finds one? Automatically creates a Notion task with context. I never miss a deadline and I get my mental bandwidth back. Small automations compound and they’ll save you hours, even days, over the year. If you’re tired of doing the same things twice, start here → https://lnkd.in/egZAQB4z #ZapierPartner 

We don’t talk enough about the figure-it-out people. You know the ones. The folks who quietly go learn the new tool, wire up the system, and make things work before the company even realizes it needed doing. They don’t wait for permission or perfect clarity. They just start building. And weirdly, those same people are often the ones who feel the most out of place right now. They’re not sure what to call themselves. They’re not data scientists. They’re not engineers. They’re not influencers. They’re just figuring it out. Here’s the truth: That instinct to learn fast, connect dots, and turn chaos into clarity is the essential skill of the AI era. If you’re that person: • You’re not behind. • You’re not an imposter. • You’re infrastructure. The future of AI transformation won’t come from consultants in slide decks. It’s going to come from the inside. From the builders, the operators, the ones who already see how it could all work better. You don’t need to become someone else to matter in this next wave. You just need to recognize what you already are. I’m spending the next few months helping these “figure-it-out” people step into real AI Transformation Lead roles inside their companies and across industries. If that’s you, or you want to hire someone like that, let’s connect.

If you build AI agents, don't miss this session. Next Tuesday, we're going inside the RL environments that make them work. Join Will Brown (Prime Intellect) and our own Robin Salimans as they show how we design spaces where agents improve safely, efficiently, and with verifiable signals. It's a deep dive on: • Why instrumented tasks matter for feedback-driven training • How Verifiers and GEPA shape stronger policies • The tradeoffs and lessons from early experiments Join us live, Tuesday December 2nd at 1 PM ET. Reserve a seat here: https://lnkd.in/ePGx-QNj

🧠 Agentic AI systems are only as good as the environments they learn in. Join Will Brown from Prime Intellect and me coming Tuesday for a chat about RL Environments! We’ll go under the hood of how we’re building reinforcement learning (RL) environments for agentic AI systems, and how verifier-driven evaluation helps us make those systems more reliable and scalable. Here's what we'll talk about 👇 1. What are RL environments? A look at how we use controlled, instrumented settings to let agents safely learn from feedback. 2. Building RL environments at Zapier How we’re applying the Verifiers library and GEPA (Genetic-Pareto) optimization to evaluate and refine agent behavior on real automation tasks. 3. Lessons learned so far Early insights into what’s worked, what hasn’t, and how we’re evolving our approach to agent training and evaluation. We'll also have some cool surprises for you, so make sure to join! 😏 If you’re interested in RL, verifiable evaluation, or applying agentic methods in production, then this should be a good session. 🎙️ Under the Hood: RL Environments at Zapier 🗓️ Tuesday, December 2, 1 PM ET Link to register in the comments

⚡ Lead Router is task-FREE for a limited time Zapier’s Lead Router helps teams distribute leads fairly and efficiently — and right now, routing won’t use any Tasks. ✅ Ensure equitable distribution with round robin and weighted routing ✅ Keep all reps in a centralized table that Zaps can reference directly — no more extra find steps or wasted Tasks ✅ Balance workloads automatically and adjust rules anytime with a simple, no-code UI It’s a great time to see how effortless and efficient lead routing can be — while it’s task-free. Learn more: https://lnkd.in/gdS2CAYd Start building: https://lnkd.in/gxHG5QNa Build along with me as I set up a router: https://lnkd.in/gKxz3cv6

Most AI pilots never make it to production. We found out why 👀 In our new AI Execution Gap Report: 👉 74% of leaders say half or fewer pilots survive 👉 91% of practitioners say projects pause after pilots 👉 And 46% point to integration chaos as the top barrier 😬 See how top enterprises are turning ambition into results. Read the report today: