Dog Care Chatbot Development
Budget: €750 – €1,500 EUR
The project involves building a text-based chatbot that can chat naturally with our clients and give reliable, friendly advice on general dog care. I want the bot to understand free-form questions (feeding schedules, grooming routines, exercise needs, basic wellness pointers, etc.), pull the right answer from a curated knowledge base, and reply instantly through a simple text interface on our website and, later, inside our mobile app.
Core expectations
• Conversational accuracy: the NLP must recognise common dog-related phrases, misspellings, and emotive language, then map them to the correct care tip.
• Easy upkeep: I should be able to update or expand the knowledge base without touching code.
• Smooth integration: embed the chat window in our existing React site and expose a REST endpoint so our app team can reuse the same engine.
• User hand-off: when confidence drops below a threshold, the bot flags the chat and hands it to a human agent by email or Slack.
• Analytics: basic dashboard or API that shows volume, top questions, and unresolved queries.
Preferred stack is open—Python with Rasa, Dialogflow CX, or a lightweight Node solution are all fine as long as you explain the trade-offs. Model training data, deployment script (Docker or similar), and concise setup documentation are part of the final hand-over.
A brief demo on your own server first, then we migrate to our AWS account. When the migrated bot matches the demo behaviour and the above criteria, the job is done.
Core expectations
• Conversational accuracy: the NLP must recognise common dog-related phrases, misspellings, and emotive language, then map them to the correct care tip.
• Easy upkeep: I should be able to update or expand the knowledge base without touching code.
• Smooth integration: embed the chat window in our existing React site and expose a REST endpoint so our app team can reuse the same engine.
• User hand-off: when confidence drops below a threshold, the bot flags the chat and hands it to a human agent by email or Slack.
• Analytics: basic dashboard or API that shows volume, top questions, and unresolved queries.
Preferred stack is open—Python with Rasa, Dialogflow CX, or a lightweight Node solution are all fine as long as you explain the trade-offs. Model training data, deployment script (Docker or similar), and concise setup documentation are part of the final hand-over.
A brief demo on your own server first, then we migrate to our AWS account. When the migrated bot matches the demo behaviour and the above criteria, the job is done.