Caddi

Everyone's building AI for the expert, client-facing work. We're automating the other 50%. The back-office. No coding. No prompt engineering. Your non-technical team just hops on a call, shows the AI how they run intake, conflicts, or pre-bill review, and the agent learns to do it for them. Not a chatbot. An AI workforce that makes your team 10x more effective. Listen to our CEO Alejandro Castellano give a 2-min overview on how:

The opportunity in operations isn't another tool. It's getting the tools you already have to work like one system. For Matthew Mercer, COO of The Planning Center, Inc., a fee-only RIA with five offices from Alaska to New Orleans, that was the whole problem. The firm's processes were finely tuned, but every one of them ended with a person re-keying data by hand, with no developers on staff to bridge the stack. Today Caddi is a core part of how the firm runs. Roughly half the team now records their own processes and hands the busywork to Caddi, freeing them up for the work they actually want to be doing. "I don't need another tool for my clients. I need something that helps our internal lives easier and the systems we have talk better together." Watch the conversation: https://lnkd.in/gZ6yKu2d

Automation is reliable but rigid. AI is intelligent but hallucinates. Agentic Automation combines the reliability of automation and the intelligence of AI. #agenticautomation #ai #legaltech #ria #professionalservices

Law firm profits grew 13% last year. So why does every managing partner I know feel margin slipping through their fingers? Three forces converging at once: - 85% of clients now demand pricing transparency - AI is compressing the work, but the billable hour hasn't moved - Rate pressure keeps climbing You cannot price your way out of this if you don't know your true cost-to-serve. Defending margin is a data problem long before it's a pricing one.

Tech-stack sprawl, a compliance bar that keeps rising, and the M&A and succession crunch. The three operational challenges defining RIAs in 2026, why they all land on the COO, and how the firms pulling ahead are answering them.

"But what if the AI hallucinates?" Right fear. Wrong target. The problem was never automation itself. It's using a generative model that guesses, in places where a wrong guess goes unreviewed. The answer is deterministic, code-based execution. Software that runs the same verified steps every time. No guessing means nothing to hallucinate. Here's why that distinction is everything for back-office work:

About 36% of the top 2,000 US law firms are now using AI in production, not just piloting it. The real story is how broadly it has spread: across six very different kinds of work, from drafting to the back office.

Over $10B in AUM, the question is leverage: how to add advisors and acquire firms without adding ops in lockstep. The best tool scales advisor support without scaling headcount.

"Show me the ROI." That's where most law firm AI projects go to die. Not because the value isn't real, but because it's buried in non-billable work nobody ever tracked. You can't prove savings on hours you never measured. So we wrote the actual business model. The four numbers a managing partner will sign off on: - Non-billable hours recovered - Cycle time cut - Write-downs avoided - Capacity added Baseline those four and you have a case the finance committee can't wave away.

Strip away the trend lists and the same three problems sit on every legal ops leader's desk this year: proving the payoff on AI spend, holding on to the people who run the operation, and defending margin while clients demand price transparency. They look like three problems. They share one root cause.