Hashlogics
← Case studies
Case study · Travel & Hospitality · United States

Cruise Search AI

Find the right cruise by just describing it.

AI cruise discovery for military and veteran travelers.

Cruise Search AI product interface
Client
Military Cruise Deals
Industry
Travel & Hospitality
Region
United States
Engagement
B2C · assisted cruise booking · Embedded AI search module
Overview

We built an AI cruise-matching module for militarycruisedeals.com, the client's existing WordPress site. Instead of scrolling long listings, travelers describe what they want in plain words. The AI asks a few short follow-up questions, turns the request into validated search filters for ships, ports, and dates, and returns one clean results link to the client's own search engine.

The challenge

The problem we set out to solve.

01

Customers searched in vague, conversational ways and changed their criteria midway.

02

Long, scroll-heavy listings made the site hard to use.

03

Mismatched filters produced false "no results" pages that lost bookings.

04

Date handling had to follow strict departure-window rules to stay accurate.

What success needed to look like

  • Let customers search the way they naturally talk, not fill out a rigid filter form.
  • Stop false "no results" outcomes caused by mismatched filters.
  • Hand back one clean results link instead of a long, scroll-heavy list.
  • Give the team visibility into what customers are actually searching for.
Our approach

How we delivered it.

01

Diagnose

Traced real search sessions to find where vague requests, mismatched filters, and session carryover were producing false "no results" outcomes.

02

Design

Designed the clarifying-question flow and the rules for turning conversational intent into validated ship, port, and date parameters.

03

Build

Built the FastAPI and LangGraph agent with Redis session state, then embedded it as a widget inside the client's existing WordPress site.

04

Launch

Connected the chatbot to the client's existing cruise search engine so every conversation ends in one clickable results link.

05

Run

Added an admin dashboard for reviewing recent searches and put controls in place to keep OpenAI usage reliable.

The solution

What we built.

We built a conversational cruise-search module inside the client's WordPress site. It asks short clarifying questions, then translates intent into validated parameters for ships, ports, and dates. It checks each request against real ship, port, and date combinations, so travelers get a single clickable results URL that points to the client's existing search engine. An admin dashboard lets the team review recent searches, and controlled AI usage keeps costs and reliability in check.

Conversational cruise search with guided refinement
Exploratory handling for vague or undecided requests
Natural language turned into validated search filters
Single clean results URL to the existing search engine
Ship-aware handling to cut unnecessary follow-ups
Validation against real ship, port, and date combinations
Admin dashboard for reviewing recent searches
Military Cruise Deals homepage with the cruise search panel
Cruise deals browsable by destination
How it’s built

Tech stack

WordPress (custom module)FastAPILangGraphRedisAWS EC2

Integrations

  • OpenAI API
  • Client's cruise search engine
More from the build
Cruise line partners and the military discount booking guide
Military Cruise Deals search experience on a laptop
The takeaway

A good search assistant doesn't just answer questions — it turns a vague, conversational request into a precise, valid search every time.

Start

Let’s build the one that runs after.

A senior engineer reads every brief, and your call gets scheduled within 24 hours.

What happens next

  1. 01

    You send a brief or book a call

    Two minutes, whichever you prefer.

  2. 02

    A senior engineer replies within 24 hours

    Not a sales rep.

  3. 03

    Honest scoping, in writing

    And if we’re not the right fit, we say so.

Abdul Basit, CEO of Hashlogics

“I started Hashlogics because too many teams ship a demo, get paid, and disappear. We build to a standard we’d run ourselves — and we stay to keep it running.”

Abdul Basit · CEO · a direct line

Not ready to talk? Take the checklist.

12 questions to ask any AI agency before you sign — the ones that separate a demo shop from a team that ships to production.

Get the checklist

Free · no newsletter