Loopcall: Meet Sonali, a Voice - First AI Agent that Confirms D2C Orders At Zero Cost

Description and use

Loopcall automates order verification using a local Excel workbook and open‑source models, reducing manual calls and preventing losses from fake COD orders.

Problem

Incomplete delivery details and fake Cash on Delivery (COD) orders cause manual calls, delayed fulfilment, and financial loss for small e-commerce businesses.

Outcome

Loopcall reduced manual calls, prevented failed deliveries from fake COD orders, and maintained order data locally without external cloud services.

Workflow

• Read flagged order from local Excel workbook
• Identify concern (e.g., missing pincode, unverified COD)
• Initiate outbound call using a telephony provider
• Guide customer through a concise, contextual conversation
• Analyze response to determine verification success or need for human review
• Update relevant fields in the original Excel workbook
• Flag complex cases for human intervention

Evidence

Incomplete delivery details mean more manual calls, delayed fulfilment, and wasted operational time.

Source

Sonali reads order details from a local Excel workbook, identifies the flagged concern, and calls the customer to resolve it through a natural, conversational voice interaction.

Source

Loopcall uses an offline Excel workbook as its source of truth.

Source

Three open-weight models on one laptop: Whisper-Hindi2Hinglish-Swift, Gemma 4 E4B via Ollama, Kokoro-82M.

Source