Wade Foster

A product's ready to ship, but it'll be 10x better in a month. Do you wait?? My answer: ship now. Always. The product that exists beats the one you're still imagining. @Wistia CEO @CSavage agrees. He doubled his team, but found he shipped slower. So he killed centralized roadmaps and replaced them with one rule: own the customer problem + ship every 2 weeks. 12 updates a year became 70, then 100+. Same team size. Listen to my full convo with Chris here: youtu.be/LHd1plfMvXA?si…

There’s no such thing as institutional AI yet, according to @a16z. They're right. For now. Quick history lesson: when textile mills went electric in the 1890’s, they just swapped steam engines for electric motors. Same layout and workflow Took 30 YEARS before they increased output bc no one thought to redesign the factory... That's where most companies are with AI right now. Individual employees are faster, electrified. But the org? Stuck in the steam engine days. At @Zapier, we have ~800 people using AI daily. Many of them have built personal operating systems that have 5x’d their individual work. How do you scale that across hundreds? That’s the question I’m obsessed with. Any wild experiments out there?

This guide changed how I work every day. It was like a gateway drug. tl;dr: Instead of designing a folder structure, @Obie just said "build me a markdown-based OS that makes me a world-class CTO" My personal AI system is now built on this idea. CRM, daily briefing, and my personal ‘war council’ for planning. It started as one prompt and now it runs my week.

More ICs are choosing to stay ICs. Our CTO @BryanHelmig is a great example. He prototypes, ships features, writes code, teaches, and leads. He just doesn’t have direct reports. It’s been that way for a long time now (even further back than this pic 😂) I'm seeing more of this at Zapier - people who could move into management and are choosing not to. They'd rather manage AI agents and build. So we're leaning in. More builders, fewer layers. Giving motivated ICs unlimited access to AI tools. The rise of the Super IC is here. We’re setting @Zapier up to attract the very best. Sound fun? We’re hiring (~30 open roles and counting…) zpr.io/ASYpwVuiyLMT

Nobody knows great talent like your own team So we made it as easy as possible for every employee to make a referral 👀 We built an AI agent that turns referrals into a real pipeline: • submitted in seconds • instantly actionable by recruiters • tracked cleanly in the ATS Submit a form in Slack, Zapier handles the rest See it in action with Emily, one of our Sr. AI Automation Engineers:

Meetings are expensive. Make decisions from prototypes, not slide decks. This matches what I've seen the best teams do at @Zapier. For decades, orgs were structured around code being expensive, slow, and hard to reverse. The move now is to eliminate planning for anything you can test empirically. Build in 2 hours -> measure -> kill or keep May the best build win.

The best founders I know are using AI to replace prep work. @ClaireVo calls hers the "Sunday Scaries" agent. No code, just @Zapier and a prompt. Every Sunday it reads her calendar, looks up who she's meeting, and finds the mornings she can drop her kids at school. It was built in an afternoon. Her engineering team is herself and one full-time engineer. But she's shipping features before lunch. Her advice for leaders who feel like they’re falling behind: • Build one agent. Start with something like her Sunday Scaries flow • Vibe code a feature you wish existed • Run an exec hackathon (bonus points if it's for the leadership team) Watch the full episode here: youtube.com/watch?v=_Wg2oT…

The real estate industry still reviews leases by hand. So one CFO automated it himself. I asked Carlos Olea (CFO of @HowardHughesHQ) what he'd do if someone automated themselves out of a job... "I'd have them automate everything else" New episode of Agents of Scale here: youtube.com/watch?v=W9IzhS…

We got early access to GPT 5.4. It's the new state of the art for multi-step tool use. Now available in Zapier. We run some of the most rigorous tool use benchmarks in the industry, testing models across hundreds of advanced real-world workflows. The results: @OpenAI's 5.4 finished the job where previous models gave up. The most persistent model to date. Here's the pattern we kept seeing: prior models would attempt a task 3 to 5x before quitting. GPT 5.4 might try 10, 20, even 40 times. (AKA if you've ever caught yourself saying "I promise it's there, keep looking" to your AI, that's the gap 5.4 closes) Plus? A 1m context window. Huge. Only one tradeoff: that persistence means it can be overeager. It moves fast, so build a plan first. All told, this is a model you want in your automation stack. It balances persistence with clean execution, chaining outputs across multiple tools without losing the thread. It's built for complex operations workflows where success depends on getting every step right. Give it a try in @Zapier and let us know what you're seeing.

The AI feature in @zapier is bloody awesome. I would have no idea how to do most of the things it does. Saves hoursss

A timeless tweet.

We surveyed 200 enterprise leaders about AI. 83% said their goals were realistic. But only 26% actually scaled them. @Zapier and our AI Leaders Lab built a framework to close this gap: Leadership x Culture x Tools x Governance = Impact Remove any variable and the equation collapses. See full framework, research, and a self-assessment checklist here: zpr.io/wLe8tFxrveMU

Join the most exclusive camping experience in the country. Applications are now open for @Zapier Outpost ‘26. It's the private retreat for leaders shaping AI transformation. Attendance is extremely limited (by design) Request an invitation here: zpr.io/rvp3qpdPX4V8

This guy went from living in his car to a $1+ million AI automation agency. The business is super simple. Instead of over-engineering with Claude Code, OpenClaw, etc he often uses @zapier zaps. I was skeptical. So he screenshared and showed me what his clients pay him for. (YouTube link + full templates in the first comment)

Is adulting fun now? I built a second brain for my health. On a Saturday. How I did it: Sat down in Cursor, dumped in everything I had. Lifting data, smart scale numbers, 23andMe genome, and more. Simply told it: ask questions to fill in the gaps. Same playbook I used to build my 'work brain' From there it asked things I hadn't thought about, and built out a full health profile and playbook. Now every Saturday morning I check in: the good, the bad, the ugly. It updates the playbook and tracks what's trending. Whole thing took an afternoon to build. Next up: blood panels. (Was going to share a screengrab of my dashboard but not trying to be Bryan Johnson here 😅)

Sonnet 4.6 is here. Less than two weeks after Opus 4.6. Does it live up to the hype? Yes it does. We ran @Claude Sonnet 4.6 head-to-head with 4.5 and Opus 4.6 on our Workflow Automation Benchmark. On calendar-CRM coordination and conditional scheduling routing, Sonnet 4.6 was the only one that executed correctly end to end. The kind of task where the model has to evaluate conditions, pull the right records, and pick the right path…even a 10% error rate could lose a deal. Overall, Sonnet 4.6 is a clear step up from 4.5, uses fewer tokens, and lands close to Opus performance in our testing. So when would I prefer which? I'd start with Sonnet 4.6 for workflow automation: conditional routing, calendar-CRM coordination, document and contract flows. Across our benchmarks the reasoning was close enough to Opus on those tasks, with far fewer tokens. That's my default for most Zaps. I'd reach for Opus 4.6 when the workflow needs code generation or heavy coding support, when it spans a lot of tools and the model has to synthesize across several sources, or when steps can fail and you need the model to recover and try a different approach. Opus still has the edge there. We shared these results with @AnthropicAI, and are excited to keep partnering on future launches so we can help more shared customers get the most out of Claude inside @Zapier. Learn more here: zpr.io/pBHPReyxnnjM

Got to hand it to @zapier and their integration of copilot for writing zaps for you. They seem to have thought hard about where it gets it's info from (the actual documentation of the things you are using, and not some random web search) - as a result, it actually does a good job

Bloomberg asked me about AI workers making $1,000 an hour. "Can it last?" My take: training AI pays well today. Orchestrating AI will pay even better tomorrow. Most companies hire people to label data, then can't figure out how to ship. Hybrid workflows (AI + deterministic steps) win on cost and trust. Read the full @Bloomberg article here: bloomberg.com/opinion/articl…

We are so back. @X integration back on Zapier.

how to use Zapier to easily connect just about any data/tool to Claude Code -create stunning reports - automate social content - sky’s the limit you can access 8000+ apps