Our work in AI Implementation & Engineering

Trexin built an agentic prospecting engine on Claude for Freese Project Solutions, a Twin Cities commercial real estate services startup: 10 qualified deal signals and an outreach-ready broker introduction on day one, for under $60 a month.

Trexin lift-and-shifted a fintech platform into AWS and made it U.S. production-ready in four weeks, giving a VC group a wealth-management product it fully owned, and a first-mover advantage.

Trexin replaced a costly third-party PSA SaaS with a custom Microsoft 365 app it built in three months, recapturing 100% of the annual subscription spend (over $100k) with no added licensing cost.

Trexin built an agentic AI proof of concept: a Voice AI Agent that creates, changes, and cancels airline reservations through natural spoken conversation, via an MCP server over a legacy GDS. The Client adopted it to Beta GA.

Trexin managed a Microsoft Azure + UiPath document-automation pilot for a large health insurer, projected $1.2M in first-year savings at a success rate 12% above target.

A large health insurer took back control of an underperforming, third-party AI prior-authorization model, and Trexin built the strategy and roadmap behind $6.25M in annual savings.

Our CTIO on a real, measurable LLM use case: turning millions of dollars of manual document data entry into a controllable, multi-model extraction pipeline, and why the right answer often isn't the biggest model.

A fast-growing senior-care provider's PACS imaging wasn't integrated with its EMR, hampering radiologists. Trexin closed the gap with RPA (UiPath + HL7): faster image access and a 20% increase in imaging throughput.

Growing 35%+ a year and opening 20 clinics, a senior-care provider needed automation to scale. Trexin's 8-week RPA pilot automated over 90% of tasks, cut errors 70%, and projected $2M+ ROI.

A malpractice insurer estimated $4M a year from catching high-risk claims at first notice of loss, but review was manual and expert-bound. Trexin's machine-learning model learned the rules from history; within a year the Client was on track to take down $4.3M.

A state Medicaid plan set a $4MM cost-of-care savings target. Trexin's machine-learning model learned the experts' member-reassignment rules at over 99% accuracy, cutting cycle time an estimated 8–12 weeks and driving more than $1MM of the goal.