← All work

Winemaker Travel / Web

A clear path from a sales request to a winery visit.

Full-stack engineering for Constellation Brands' internal platform coordinating requests, approvals, winery visits, and feedback.

My role
Fullstack / AI Engineer Intern
When
June 2026 — present
Built with
Next.js · NestJS · MySQL · CQRS · Docker · GitHub Actions · Spec Kit · GitHub Copilot
The original Winemaker Travel operations queue showing sample requests, review status, and coordination tools
Winemaker Travel · actual application running locally with synthetic visit records · Open full size ↗

Explore how it works

Try it yourself ↓
Interactive explanation · synthetic sample data
CONSTELLATION BRANDS / WINEMAKER TRAVEL

Keep every handoff tied to the visit.

Step through a sample request, from review to the final recap.

  1. ReviewOperations
  2. AvailabilityTraveler
  3. AcceptedTraveler
  4. ConfirmedOperations
  5. RecapsSales + Traveler
  6. ClosedOperations
SAMPLE TRAVEL RECORDv1

Market visit

14–15OCTOBER · SAMPLE DATES
Request
Sales team
Coordination
Operations
Traveler
Sample winemaker

Try changing the dates after acceptance or confirmation.

Operations

A shared request, ready for review.

Operations reviews the visit and requests availability from the assigned traveler.

Illustrative interface based on the implemented visit lifecycle. Each click simulates a role’s action; completion assumes the visit has ended. No real trip, email, calendar entry, or feedback is created.

The challenge

Coordinating sales requests and winery visits requires a consistent workflow across approvals and follow-up. In parallel, AI-assisted implementation needed clear rules linking specifications, plans, code, and review.

How I built it

Tie acceptance to the current proposal

I built a coordinated request lifecycle across Next.js, NestJS, and MySQL. Traveler acceptance belongs to a specific proposal and assigned person; material changes require a fresh response before Operations confirms the visit.

Make the delivery process explicit

I built a Spec Kit / GitHub Copilot governance framework with constitution checks, prompt gates, protected policies, and SHA integrity validation across the specification → planning → implementation → review lifecycle.

Close the loop deliberately

I implemented separate Sales and Traveler recap assignments and explicit closure checks, supported by optimistic concurrency and transactional updates. I also defined a CQRS / BFF target architecture for future service extraction.

What came out of it

Full-stack development and an AI-governed delivery framework for an internal platform at Constellation Brands. The work connects application architecture with practical controls for how the software is built.

The example uses a fictional visit. It illustrates application behavior without displaying internal trips or sending email and calendar invitations.

Sources & project context

More from the project

Winemaker Travel’s original visit detail showing a confirmed sample proposal, follow-up actions, traveler information, and readiness checks
Winemaker Travel · original visit detail with synthetic people and dates · Open full size ↗
Next projectKia waste dashboard ↗