Email Sales Auto-Responder AI
Budget: €250 – €750 EUR
I need a lightweight AI agent that can sit behind my company’s shared inbox and automatically draft replies to incoming sales inquiries. The interaction channel is email only, so there’s no requirement for live chat widgets, mobile SDKs, or multichannel routing—everything happens through standard IMAP/SMTP or an email API such as Gmail, Microsoft 365, or SendGrid.
The scope is intentionally simple: the bot just has to spot key phrases and product names in the message body, match them to a predefined response, then return a polished reply that I can approve or that can be sent automatically. We are not aiming for deep natural-language reasoning; a well-structured keyword or rule-based engine is fine as long as it is easy for me to update when new promotions or product lines launch.
Key deliverables:
• A small, self-contained script or microservice (Python preferred, but I’m open to Node.js or similar) that logs into the email account, scans unread messages, and produces the reply text.
• A simple configuration file or dashboard where I can add, edit, and prioritise keyword-to-response pairs.
• Clear instructions for deploying the service on a modest VPS or serverless function and scheduling it to run at regular intervals.
• A brief test suite or sample email set that proves the matching logic works and no genuine message is left unanswered.
Acceptance criteria:
1. When a test email mentioning a listed keyword arrives, the system generates the mapped response within 60 seconds.
2. Emails without any recognised keyword are flagged for manual review rather than receiving a blank or incorrect reply.
3. I can add a new keyword and response through the config, restart the service, and see it take effect immediately.
If you’ve built rule-based autoresponders before and can get a prototype running quickly, let’s talk.
The scope is intentionally simple: the bot just has to spot key phrases and product names in the message body, match them to a predefined response, then return a polished reply that I can approve or that can be sent automatically. We are not aiming for deep natural-language reasoning; a well-structured keyword or rule-based engine is fine as long as it is easy for me to update when new promotions or product lines launch.
Key deliverables:
• A small, self-contained script or microservice (Python preferred, but I’m open to Node.js or similar) that logs into the email account, scans unread messages, and produces the reply text.
• A simple configuration file or dashboard where I can add, edit, and prioritise keyword-to-response pairs.
• Clear instructions for deploying the service on a modest VPS or serverless function and scheduling it to run at regular intervals.
• A brief test suite or sample email set that proves the matching logic works and no genuine message is left unanswered.
Acceptance criteria:
1. When a test email mentioning a listed keyword arrives, the system generates the mapped response within 60 seconds.
2. Emails without any recognised keyword are flagged for manual review rather than receiving a blank or incorrect reply.
3. I can add a new keyword and response through the config, restart the service, and see it take effect immediately.
If you’ve built rule-based autoresponders before and can get a prototype running quickly, let’s talk.