About Corvino Labs
EPL Media had reached the point where more revenue meant more people. We shadowed the business, connected their systems, and took the manual steps out of the work. Their own team now extends it without us in the room.
Who we are
Corvino Labs is led by Casey Corvino, CTO and co-founder of Lavender, where he built and ran AI in production for a real company with real customers. Not a demo. That’s the whole basis of how we work.
We’re not here to run you a workshop on prompting. We’re here to make the thing work for every person you employ, then hand it over so your team can run it without us.
You are not sold by a partner and then handed to somebody three years out of school. The people who shadowed EPL’s business are the people who built it and trained their staff on it.
Where we learned it
Corvino Labs did not start with a thesis about AI. It started with a decade of shipping software under the two constraints every company has: not enough people, and customers who notice immediately when something breaks.
Casey finished New York University in three years and went straight to CTO of an early-stage startup automating fashion PR.
It set the pattern for everything after. A team too small to specialise, a budget too thin to buy your way out of a problem, and software that had to work on Monday morning.
He led engineering on Sorter alongside Will Allred and William Ballance, applying personality and communication psychology to marketing.
The lesson that outlasted the company: the hard part of communication software is never generating the words. It is knowing what good looks like, and saying so before the message goes out.
The three of them rebuilt it around one sharp job. Coach a salesperson on the email they are writing, in real time, while they can still change it.
Casey co-founded Lavender as CTO, owning product and engineering.
Lavender grew past 30,000 active users across Twilio, Clari, Sendoso, Demandbase, Segment, MasterClass and thousands of smaller teams.
It raised an $11M Series A led by top-tier investors including Norwest Venture Partners.
Casey stayed customer-facing the whole way, hyperfocused on the specific problem in front of the specific customer instead of the roadmap in the abstract.
Being CTO was not just building the product. It meant getting the rest of the company to actually adopt AI and get 10x out of it, not just the engineering team.
Companies kept asking a version of the same question: we bought the AI, why is nothing different?
The answer was always the same three gaps, and we had already spent years closing them for ourselves. So we started a company to close them for other people’s teams.
What we built there
Lavender was the thing customers paid for, running every working day on their real email. Four decisions made that possible, and all four are load-bearing in what we install for you.
Lavender did not wait in a separate tab to be visited. It worked on the draft in front of the rep, at the only moment the advice was still worth anything: before they hit send.
AI that requires somebody to remember to go and use it gets used by the people who were already enthusiastic. That is the same three volunteers every time.
Every email scored against what the system had learned from millions of real ones, turned into specific fixes rather than a grade. That is Sorter’s inheritance, made operational.
Lavender was never a generative AI company. It was a communication intelligence company that happened to write text at the end.
Casey adopted large language models early, well before it was fashionable, and made a deliberate call not to staff a research team to do it.
Lavender hired backend engineers to use the models that already existed. It saved millions and kept the product cheap enough to sell. It is why we will never try to sell you a model.
Almost none of the difficulty was intelligence. It was access, plumbing, and the unglamorous integrations that had to hold while thirty thousand people worked on their own live data.
Running AI for that many people is a different discipline from demonstrating it once. That difference is most of what we sell.
“I was an early adopter of AI. Thousands of hours later, I know what actually works for a scaling org and what doesn’t. Now I’m putting it to work for you.”
Casey Corvino, co-founder & CTO, Lavender
Thirty minutes. Tell us what your team does all day and we’ll tell you honestly whether we can help, including when the answer is no.