AI-Powered Customer Support Workflow (Concept Design)
A Miro-designed workflow concept for cutting hold times to under 5 minutes by routing routine tickets through AI before they ever reach an agent.
Six production-style projects across conversational AI, RPA, predictive automation, and no-code orchestration — each one solving a real business problem end-to-end.
Each project ships with the problem, the solution, the tools used, and what I'd improve next.
Projects 1–7 were built as part of TripleTen's AI Automation course.
A Miro-designed workflow concept for cutting hold times to under 5 minutes by routing routine tickets through AI before they ever reach an agent.
A Zapier-built chatbot that answers history students' questions from a primary source — with proper Chicago-style citations.
Sentiment-tagged customer reviews flow into a spreadsheet automatically — and negative ones fire an instant alert.
Every morning, a beachside kiosk's weather, staffing plan, and social copy land in the manager's inbox — automatically.
A UiPath bot that reads invoices, extracts the key fields, and logs them to Google Sheets without a human in the loop.
Built from real-world frustration: a workflow that predicts delivery truck arrivals so a merchandiser never wastes a morning waiting again.
Two conversion-focused landing pages — a dog grooming spa and a home cleaning service — designed to turn local search traffic into booked appointments.
Custom skills I built for Claude — extending Anthropic's skill format to handle real workflow gaps I hit in personal use.
Projects I built outside of client and platform work — shipped, tested against ground truth, and written up honestly.
Prompt-engineered NCAA tournament prediction system that hit 73% game accuracy across 63 games of the 2026 tournament — within 2 points of the published academic benchmark.
Adapted from: Foundation methodology — Bart Torvik's tournament analytics framework. Kill-shot margin metric — EvanMiya. Trapezoid of Excellence — Ryan Hammer.
I'm Trey — a TripleTen AI Automation graduate who builds workflows that take repetitive, time-draining tasks off people's plates so they can focus on the work that actually requires a human.
My approach is shaped by my day job as a retail merchandiser: I see firsthand where small inefficiencies steal entire mornings, and I build automations that close those gaps. Several of my projects come straight from problems I or my coworkers hit every week.
I'm comfortable across no-code platforms, RPA tools, and modern AI APIs — and equally comfortable jumping into a new tool to ship the right solution.
I'm actively interviewing for AI automation, RPA, and workflow engineering roles. Pick whichever channel works for you.