---
title: "Explaining Every Rupee: How We Built Reliable LLM Support Bots for Delivery Partners"
description: "Swiggy faced close to 9 lakh monthly payout- and incentive-related tickets, where delivery executives needed accurate explanations of earnings, deductions, missed incentives, and credited amounts. The system separates financial truth from language generation: deterministic preprocessing and internal APIs resolve incentive eligibility, payout breakdowns, deductions, and conditions before an LLM explains the result. Two multilingual bots—the Daily Incentive Bot and Payout Bot—use a LangGraph StateGraph with LLM, tool, and validator nodes, Redis checkpointing for multi-turn context, and post-processing translation into the executive’s preferred language. Azure Translator preserve-list controls help keep business-critical terminology consistent, while structured escalation sends out-of-scope cases and context to human agents. Moving retrieval from Databricks SQL queries to an aggregated internal API reduced latency from 8–10 seconds to milliseconds; the bots now resolve about 90% end to end, cut resolution from roughly 20 minutes to seconds, and improved CSAT by 40%."
---

# Explaining Every Rupee: How We Built Reliable LLM Support Bots for Delivery Partners

[Swiggy](https://yomu.fyi/company/swiggy) · Yash Khandelwal · May 29, 2026

**Type:** Problem & solution

## Summary

Swiggy faced close to 9 lakh monthly payout- and incentive-related tickets, where delivery executives needed accurate explanations of earnings, deductions, missed incentives, and credited amounts. The system separates financial truth from language generation: deterministic preprocessing and internal APIs resolve incentive eligibility, payout breakdowns, deductions, and conditions before an LLM explains the result. Two multilingual bots—the Daily Incentive Bot and Payout Bot—use a LangGraph StateGraph with LLM, tool, and validator nodes, Redis checkpointing for multi-turn context, and post-processing translation into the executive’s preferred language. Azure Translator preserve-list controls help keep business-critical terminology consistent, while structured escalation sends out-of-scope cases and context to human agents. Moving retrieval from Databricks SQL queries to an aggregated internal API reduced latency from 8–10 seconds to milliseconds; the bots now resolve about 90% end to end, cut resolution from roughly 20 minutes to seconds, and improved CSAT by 40%.

## Context

Delivery executives needed fast, accurate explanations for incentive eligibility, payout differences, deductions, taxes, onboarding fees, penalties, and final credited amounts. Resolving these financial-support questions required cross-referencing complex conditions and data distributed across multiple systems, creating long resolution times, low CSAT, high human-support volume, and a trust risk when answers were wrong.

## Approach / What changed

The system uses deterministic logic and internal APIs to resolve payout and incentive facts before an LLM generates an explanation. Daily Incentive and Payout bots are implemented with a LangGraph StateGraph containing LLM, tool, and validator nodes, with Redis checkpointing for multi-turn conversations. Translation is applied after generation, business terminology is controlled through Azure Translator preservation mechanisms, and out-of-scope cases escalate with conversation context to human agents.

## Takeaways

- The validator compares claimed incentive amounts in the generated response with raw tool output and sends mismatches back for regeneration.
- Replacing Databricks SQL retrieval and heavy joins with an aggregated internal API reduced data-fetch latency from 8–10 seconds to milliseconds.
- The bots resolve about 90% of incentive- and payout-related queries end to end, reduce resolution time from roughly 20 minutes to a few seconds, and improve CSAT by 40%.

**Tags:** [Databricks](https://yomu.fyi/topic/databricks), [LangGraph](https://yomu.fyi/topic/langgraph), [LLMs](https://yomu.fyi/topic/llm)

- Source: [Swiggy](https://medium.com/swiggy-bytes/explaining-every-rupee-how-we-built-reliable-llm-support-bots-for-delivery-partners-066c4f23e875)
- Source URL: https://medium.com/swiggy-bytes/explaining-every-rupee-how-we-built-reliable-llm-support-bots-for-delivery-partners-066c4f23e875
- Ingested by Yomu: 2026-09-09T13:56:50.107Z

[Read original post](https://medium.com/swiggy-bytes/explaining-every-rupee-how-we-built-reliable-llm-support-bots-for-delivery-partners-066c4f23e875)