How Trellist built an AI agent on Amazon Bedrock that diagnoses a distributor's missing-product reports in plain language, fixes the one safe case on its own, and proves every fix before reporting it.
- Client
- Promotional products and branded merchandise distributor
- Industry
- Retail & Consumer Goods
The challenge
This distributor of promotional products and branded merchandise runs a catalog of hundreds of thousands of parts, fed from supplier systems on a five-minute sync. When one of those parts quietly failed to land, nothing flagged it. The first sign was usually a customer who could not find a size or color they expected.
Every report started a manual investigation. Customer service wrote a ticket, the ticket waited in a queue, and an engineer queried the catalog, the search index, and the supplier feed by hand. That took days, and the next refresh often cleared the symptom before anyone found the cause, so the same failure came back.
What we did
Trellist designed and built a catalog agent that customer service can ask in plain language. It runs inside the distributor's own AWS account, on the patterns its engineers already maintain.
- Plain-language intake. Customer service describes the problem where they already work. The agent parses the supplier and product from the text and decides which checks the description calls for.
- Six tools, five of them read-only. The agent can read what the catalog is serving, what the product system says is approved, how parts are grouped for display, and what the supplier offers live. Its one write tool rebuilds product groups, scoped to a single index.
- Claude on Amazon Bedrock. The agent reasons with Claude Sonnet 5 on Amazon Bedrock, behind API Gateway, Lambda, SQS, DynamoDB, and CloudWatch. The model can be swapped later and the tools stay in place.
- Answered in the same conversation. Every request is authenticated at the door and acknowledged right away, and the answer comes back in the same thread when the work is done.
- A full audit trail. Every diagnosis is logged with who reported it, which tools were called, the state before and after, and the verdict that came back.
How the agent decides
The agent classifies what it found before it changes anything. Only one of its four verdicts is allowed to write, and when it does, the agent re-reads the catalog to prove the fix worked before it calls the issue fixed.
The impact, measured
Missing-product reports now get a verified answer in about 15 minutes, and the agent resolves up to 70% of them without an engineer.
- 958 silently dropped products found in just five days of logs, traced to 1,093 failed writes that nothing retried and nothing alerted on.
- Every diagnosis on the record, with who reported it, what was called, the state before and after, and the verdict.
- Causes stay on the record, so a refresh can no longer hide the failure behind a report.
- Nothing new asked of customer service. They report issues the same way they always have.
Let’s talk about the investigation eating your engineers’ week
Tell us where reports pile up and we will scope an agent that reads first, acts within tight limits, and proves every fix, the same way we did here.
Required
Trellist uses the information you provide to contact you about relevant content, products, and services. You may unsubscribe at any time.