Encore AI raised a $30M Series A, ARR up 5x in 18 months

Encore AI just closed a $30 million Series A after growing annual recurring revenue fivefold in eighteen months. They sell voice and chat agents to banks, insurers, and other regulated enterprises, more than forty customers so far. The hook is simple: the AI learns from recordings of your best call-center reps, mines the patterns that actually close deals or solve problems, then replicates them at scale.

They call it Interaction Mining. Instead of training a generic chatbot on the internet or a vendor’s idea of “best practice,” Encore builds agents on transcripts and metadata from the calls that worked. A bank’s top mortgage specialist handles objections a certain way, sequences questions in a proven order, knows when to escalate. The AI watches, learns, then handles the next thousand calls the same way. It’s not magic. It’s operational data turned into repeatable process, which is exactly what enterprises will pay for.

The revenue curve and customer list prove this isn’t a science project. Regulated industries move slowly, but they moved here because the value is legible: lower handle time, fewer escalations, consistency across shifts and geographies. It’s also proof that “train the AI on what your business already does well” is a fundable, scalable wedge. That framing sits at the core of tools like the Omni Command Centre, where the goal is always to start with the operations you’ve already validated, not a blank slate.

Encore is enterprise-focused, so the deal cycles are long and the price tags are high. But the underlying lesson scales down: if you can identify the workflows in your company that actually perform, you can teach an agent to replicate them. The hard part isn’t the AI. It’s knowing which patterns are worth copying in the first place. Encore’s Series A says the market will pay handsomely for whoever figures that out. You can read more in the TechCrunch report.