Contact person: Business Client

Phone:Show

Email:Show

Location: Bengaluru, India

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I need a developer who can build a WhatsApp-based Q&A assistant that lets users ask questions about a collection of interfaith manuscripts and receive accurate answers in Hindi, English, and Urdu. The corpus is already digitised; what I’m missing is the conversational layer that recognises queries in any of those languages, pulls the right passage, and replies in the same language.

The flow I picture is straightforward: a user sends a message on WhatsApp, the bot detects the language, runs the question through a lightweight NLP pipeline (I’m open to Dialogflow, Rasa, LangChain, or your preferred stack as long as it handles multilingual intents reliably), and returns a citation or short summary drawn from the manuscript text. If the question is ambiguous, the bot should offer clarifying follow-ups.

You’ll be responsible for
• Connecting the bot to the official WhatsApp Business API (cloud or on-prem gateway—whichever you are set up to use).
• Implementing language detection and switching between Hindi, English, and Urdu responses without losing context.
• Indexing the manuscript text so answers come back fast, ideally under two seconds.
• Packaging the code and deployment steps so I can run the service on my own VPS after hand-over.

Acceptance will be based on a short live demo where I fire test questions in each language and see the correct passages returned." (client-provided description)