The Lazy Genius: How One RevOps Leader Built An AI-Powered Revenue Engine

I had a call with Jeffrey Ronaldi to hear about his career trajectory. He said “Maybe I’m lazy.” When he was a young kid, he seemed wired to optimize things. Instead of making 15 trips to carry things from his house to the garage, he found a wagon and made 5 trips. Much to his parents' amusement.

That instinct to find the most efficient path through any problem has defined Jeffrey's entire career. From selling coupons door-to-door in his first job, to becoming the Principal RevOps & BI Engineer at Mezmo where he is today. His impetus to always improve, stay relevant, and help revenue teams be as efficient as possible has led him to build an AI agent ecosystem that's transforming how his team works.

But it wasn't a straight path. Not even close.

The Puzzle, Not the Prize

Jeffrey's first real job was door-to-door sales. He would walk into strip mall businesses trying to sell coupon books. "I'd walk into a Best Buy and just approach employees to make a sale," he says.

When he was young, people told him he'd be great at sales. He could talk to anyone, and he was fearless. But what Jeffrey discovered early on was that he didn't love the conversation itself. He loved the puzzle underneath it.

Even before he had access to a CRM, he was thinking systematically: How many doors do I need to knock on to get someone to answer? How many answers before I get a yes?

When he moved into medical sales, he crushed it, but not because of his pitch. "I was able to figure out how to use the system in a way that made me more efficient," he explains. He turned a clunky MacDOS CRM that everyone hated into a scheduling machine, printing out his entire month's worth of appointments and map directions in advance for efficiency and accountability.

"I had the system working for me instead of working against the system."

That realization changed everything.

From Seller to Systems Thinker

His first tech job was as an SDR where he got his hands on Salesforce for the first time. "I was like, wow, this is a great thing. Spreadsheets! Databases!"

But again, what made Jeffrey successful wasn't some special magic on calls or emails. It was the meta-game. Who was winning? What were they doing? How could he replicate it faster and better?

When his company introduced A/B testing to their product, Jeffrey lit up. "Oh, this is great. Now I can test to make sure I'm actually saying the right thing to the right people?"

He broke into a completely new vertical by connecting patterns no one else saw. He became the only SDR hitting quota. When layoffs happened, they kept Jeffrey and gave him the whole world as his territory.

But Jeffrey was already thinking beyond sales. He'd started working with the Director of Sales Ops, who would pull him into conversations: "If you were to do this, how would you think about it?"

"I was like, oh, this is sales ops. I kind of like that job better than mine."

The Art of Doing More with Less

By the time Jeffrey got to his next role, he knew what he wanted a sales ops role. He began as an SDR. This SDR team had never hit quota before. Jeffrey reached 60 accepted opportunities despite a quota of only 10.

"I didn't care so much that I beat everyone by getting 60," he says. "I wanted to create a system where everyone could meet and exceed quota. That seemed way better to me than one person who could blow out the quarter. It's better to get everyone successful."

So he built the system that enabled everyone to hit quota.

This became his operating principle: Find a better way to do something and build a system that helps the team.

When the Data Was Messy, He Learned SQL

At one company, because of complex GTM motions and acquisitions, the standard playbooks weren't working. Sales reps were asking questions about product usage that couldn't be answered due to messy data.

"I started with exporting a CSV out of MongoDB and trying to force it into Salesforce," Jeffrey says. "Obviously, that wasn't working well."

So he taught himself reverse ETL. Then data warehousing. Then SQL. Then database engineering.

Even after his company hired a data engineering firm to set up the entire ecosystem—Fivetran, Snowflake, dbt—it all fell apart when the firm left. So Jeffrey took over.

"I learned how to be a data engineer. Because helping our people succeed and do their jobs required that all this work continue. So I learned, leveled up, and delivered."

Enter the AI Agent Era

The company Jeffrey is at now has had its ups and downs, but because they have Jeffrey thinking about how to build an AI-forward tech stack, they're able to weather the storms. And given his career-long instinct for systematizing everything—for finding the wheel when everyone else is carrying boxes—it was almost inevitable that when AI tools that could actually reason over messy, real-world data arrived, Jeffrey would figure out how to use them before almost anyone else.

When he started using Von, his first interaction was casual. He logged in, looked at the sample prompts, picked one, and waited to see what it would return.

"Pretty interesting," he thought. He noticed it was pulling in some outdated opportunity stages—but it was also flagging high-risk deals on its own. "I checked one of them, and yeah, it should have been closed out a long time ago. I wonder if Von can make that update?" It did.

Then he kept going.

Von as an AI Teammate

Once Jeffrey saw what Von could do, he started thinking bigger—but also more carefully. His CRM was a mess of legacy acquisitions and GTM pivots. The data wasn't normalized. Von wouldn't be able to give the right answers without understanding how his business actually worked.

So Jeffrey did what he's always done. He built the system.

Because Von has “organizational memory” and can be taught, he was able to train Von on the company's terminology and logic. Within a week, Von had become something Jeffrey hadn't expected. The knowledge that used to live only in Jeffrey's head—how deals were structured, what the product usage data actually meant, why certain stages looked the way they did—now lived in the system.

The breakthrough moment came when his sales leader, Josh, was stuck. A customer had a research project going, and Josh was drowning trying to figure out the difference between new pricing and legacy pricing. It was taking hours, maybe multiple days.

Jeffrey spent 30 minutes teaching Von about their product usage and pricing structure. Then he told Josh: "The thing you've been working on for hours? I just taught Von how to do it in 30 minutes."

Josh's response: "Woah. This is the greatest thing!"

The next time someone asked about pricing differences, or usage patterns, or how to categorize a deal—Von already knew. Jeffrey had turned a one-time fire drill into a permanent capability.

And all of that happened in Jeffrey's first week testing Von.

"It feels less like chatting with a bot and more like having an extra RevOps brain in the room," Jeffrey says. He's already planning his next set of agents—from lead qualification to message optimization to deal analysis—that can work together as a full ecosystem.

The 2026 AI-Forward Tech Stack

I asked Jeffrey to walk me through what an AI-forward stack actually looks like in practice. He's not trying to replace his systems of record—he's layering intelligence on top of them.

For data enrichment, he uses Clay. "You're able to go through a waterfall of vendors and get the number you actually need instead of going through one. When you add AI on top of it—being able to do non-standardized research like 'Have there been customer complaints about this company's app performance?'—you can't get that information anywhere else."

For internal knowledge, Notion AI has been transformative. "It integrates with Slack and Google Drive. I used it to train Von faster—I asked it to look at our board deck and documentation and build an ICP. Then I told Von, 'Commit this to your memory.' End to end, it took 15 minutes."

For CRM, he'd choose Attio if he were starting fresh. "It has a clean UI, really cool functionality, and it's intentionally more simplistic—targeted for sales, not trying to be a data warehouse." For dynamic messaging, Unify. For call recording, Fathom. For approvals, Rattle—"It's so fast, so easy, and it's one of my favorites."

And sitting over all of it: Von as the intelligence layer.

Jeffrey captured his philosophy in a LinkedIn post:

"AI products today are incredible systems of intelligence, but they are not systems of record. If an AI platform hallucinates, loses state, or misinterprets a change, your revenue numbers are wrong. That's existential risk. Where I AM seeing real value is AI as an intelligence layer on top of existing systems—not a replacement for them."

That’s what Von can do.

The Lazy Way Forward

Jeffrey Ronaldi still approaches every problem the same way he approached that wagon as a kid: What's the fastest path from A to B?

It isn’t laziness, it’s clarity and working smart. 

It's knowing that the real work isn't grinding harder—it's building systems that let people do their best work. It's not replacing humans with bots. It's giving humans superpowers.

And sometimes, that means finding the cart with wheels when everyone else is making twenty trips.

Want to see Von in action? Demo Von here or follow along on LinkedIn for more use cases. And if you want to connect with a brilliant RevOps mind, Jeffrey is happy to connect on LinkedIn.