Conversational AI Travel Assistant Demo
Budget: $250 – $750 USD
CONFIDENTIAL | AI Travel Assistant — Local Friend Persona | MVP Demo in 48h
I'm looking for an experienced full-stack developer to build a functional demo of an AI-powered travel assistant for a confidential client in the tourism sector. Full project details shared after NDA.
---
WHAT YOU'LL BUILD
A WhatsApp-connected conversational AI that behaves like a knowledgeable local friend welcoming travelers to a specific destination. Think of it as the "paisa" friend who knows every corner of Medellín, or the "chilango" friend who knows exactly where to eat in CDMX — warm, street-smart, genuinely helpful, and proud of where they're from.
The demo must show:
• AI assistant with a fixed local personality defined in the system prompt (tone, vocabulary, cultural references specific to the destination)
• Language auto-detection: Spanish, English, Portuguese, Italian, French, German
• Tone adaptation logic: detects user origin from phone prefix or message context, then adjusts register accordingly (e.g. full local slang for locals, warm neutral Spanish for other Latin Americans, friendly English for international travelers)
• At least 10 real destinations/experiences loaded into a knowledge base (RAG) — content provided by us
• Emergency mode: detects emergency keywords across all 6 languages and instantly switches to direct, no-emoji, no-slang response mode
• WhatsApp channel working end-to-end (Meta Cloud API or Twilio sandbox)
• Simple web chat fallback UI if WhatsApp setup takes too long
---
TECH STACK EXPECTED
• LLM: Claude API (Anthropic) or OpenAI GPT-4o or Equivalent
• RAG: pgvector or lightweight vector store (Chroma, Pinecone free tier)
• WhatsApp: Meta Cloud API or Twilio sandbox
• Backend: Node.js / NestJS or Python FastAPI
• Demo KB: we provide PDFs, Word files, and images for at least 10 destinations/operators
---
TIMELINE
Strict 48-hour delivery from project start. This is a demo, not production — clean code matters but speed is the priority.
---
DELIVERABLES
1. Working WhatsApp number (sandbox OK) running the bot end-to-end
2. Web chat fallback UI — simple, branded with provided logo and color palette
3. Short Loom video (3–5 min) walking through the full demo flow
4. Source code in a private GitHub repo
5. .env.example with all required API keys documented
---
YOU'RE A FIT IF YOU
• Have shipped WhatsApp bots with real LLM backends before
• Are comfortable with RAG pipelines and system prompt engineering
• Understand how to shape AI personality through prompt design (not just API calls)
• Can work fast without hand-holding
• Speak Spanish — the bot's core personality is in Spanish
---
Please include in your proposal:
→ A link to a similar project you've shipped
→ Your honest assessment of what's achievable in 48h
→ Confirmation you're comfortable signing an NDA before receiving the brief
Fixed price bids only. Full spec shared after NDA is signed.
I'm looking for an experienced full-stack developer to build a functional demo of an AI-powered travel assistant for a confidential client in the tourism sector. Full project details shared after NDA.
---
WHAT YOU'LL BUILD
A WhatsApp-connected conversational AI that behaves like a knowledgeable local friend welcoming travelers to a specific destination. Think of it as the "paisa" friend who knows every corner of Medellín, or the "chilango" friend who knows exactly where to eat in CDMX — warm, street-smart, genuinely helpful, and proud of where they're from.
The demo must show:
• AI assistant with a fixed local personality defined in the system prompt (tone, vocabulary, cultural references specific to the destination)
• Language auto-detection: Spanish, English, Portuguese, Italian, French, German
• Tone adaptation logic: detects user origin from phone prefix or message context, then adjusts register accordingly (e.g. full local slang for locals, warm neutral Spanish for other Latin Americans, friendly English for international travelers)
• At least 10 real destinations/experiences loaded into a knowledge base (RAG) — content provided by us
• Emergency mode: detects emergency keywords across all 6 languages and instantly switches to direct, no-emoji, no-slang response mode
• WhatsApp channel working end-to-end (Meta Cloud API or Twilio sandbox)
• Simple web chat fallback UI if WhatsApp setup takes too long
---
TECH STACK EXPECTED
• LLM: Claude API (Anthropic) or OpenAI GPT-4o or Equivalent
• RAG: pgvector or lightweight vector store (Chroma, Pinecone free tier)
• WhatsApp: Meta Cloud API or Twilio sandbox
• Backend: Node.js / NestJS or Python FastAPI
• Demo KB: we provide PDFs, Word files, and images for at least 10 destinations/operators
---
TIMELINE
Strict 48-hour delivery from project start. This is a demo, not production — clean code matters but speed is the priority.
---
DELIVERABLES
1. Working WhatsApp number (sandbox OK) running the bot end-to-end
2. Web chat fallback UI — simple, branded with provided logo and color palette
3. Short Loom video (3–5 min) walking through the full demo flow
4. Source code in a private GitHub repo
5. .env.example with all required API keys documented
---
YOU'RE A FIT IF YOU
• Have shipped WhatsApp bots with real LLM backends before
• Are comfortable with RAG pipelines and system prompt engineering
• Understand how to shape AI personality through prompt design (not just API calls)
• Can work fast without hand-holding
• Speak Spanish — the bot's core personality is in Spanish
---
Please include in your proposal:
→ A link to a similar project you've shipped
→ Your honest assessment of what's achievable in 48h
→ Confirmation you're comfortable signing an NDA before receiving the brief
Fixed price bids only. Full spec shared after NDA is signed.
Related categories:
Full Stack Development
Twilio
API Integration
AI Chatbot Development
Conversational AI