AI agents, RAG, and integrations — notes from production.

E-commerce

AI Support for High-SKU Retail: When Product Questions Outnumber Order Questions

For high-SKU retail, most support contacts are pre-purchase product questions - compatibility, sizing, specs - a grounded retrieval problem.

E-commerce

AI Support for Marketplaces: Handling Buyer–Seller Disputes

A marketplace support agent gathers evidence from both sides and resolves the clear cases; it doesn't adjudicate disputes, it assembles them for a human.

E-commerce

AI Support for Subscription Businesses: Cancellations, Pauses, and Save Flows

A subscription support agent handles lifecycle actions - cancel, pause, plan change - as money-adjacent write actions, plus billing questions.

E-commerce

Why an AI Support Agent Needs an Async Backend

An async framework for an AI agent matters because the workload is many connections mostly waiting - on the model, the database, a webhook - not CPU-bound work.

E-commerce

Streaming Tokens While Running a Tool Call: Keeping Support Chat Live

A streaming AI support agent sends tokens as they generate and pauses mid-stream to run a tool - so the chat forms immediately instead of freezing on a lookup.

E-commerce

One Schema Per Tool: Stopping a Malformed Refund Call Before It Runs

AI agent tool schema validation: one typed schema per tool, shared by the model spec and the executing endpoint, so a malformed refund call can't be emitted.

E-commerce

Read vs Write Tools: Designing an AI Agent's Permissions Around Money

The AI agent read vs write tools split drives the safety design: write tools get tighter thresholds, guardrails, idempotency, and an audit record.

E-commerce

AI Support Agents Beyond Live Chat: Email, Messaging, and Async Queues

AI support agent email automation reuses the chat agent, but async changes the design: one complete reply, threading, attachments, and per-channel escalation.

E-commerce

What to Log From an AI Support Agent — and Why the Audit Trail Matters

An AI support agent audit log serves three jobs: defending an action later, debugging a bad conversation, and improving the agent. What each one needs recorded.

E-commerce

How an AI Support Agent Stays Grounded in Your Help Center and Policies

Grounding an AI support agent: a relevance floor on retrieval, policy kept consistent with live order data, and why content quality decides the answer.