Reactiv cuts merchant configuration time 80% with Amazon Bedrock AgentCore
Reactiv, a provider of native mobile commerce apps for Shopify merchants, built an AI Scheduler on Amazon Bedrock AgentCore to automate app updates. Merchants can issue natural‑language requests such as “Refresh my homepage with best sellers every Monday at 9 AM,” and the three‑agent system handles intent classification, analytics, and configuration generation without manual steps.
What you can do with it
AI at Work →Update product listings daily with Reactiv AI Scheduler
Automates homepage and layout changes on a set schedule using natural-language prompts.
- Who for
- marketers, founders and online shops
- Cost
- Price not stated
- Effort
- minutes
Use it for
- Refresh homepage weekly
- Update product listings daily
- Apply brand guidelines automatically
How to set it up · From the article
- Create a scheduled task via the Reactiv dashboard or chatbot.
- EventBridge triggers a Lambda that validates the merchant and loads current config.
- Lambda invokes AgentCore runtime where three agents classify intent, analyze data, and build updates.
- Builder Agent updates the app config, stores it in DynamoDB, and awaits merchant approval.
Watch out Only available to Reactiv customers; Bedrock usage may incur extra fees
Key points from the news
- Reactiv reduced merchant configuration time by 80%, cutting 17‑hour tasks to 3 hours.
- Production rollout sped up 33% (10 weeks vs 15 weeks) after moving to AgentCore.
- Compute cost savings of about $6,000 per year and job execution halved to ~5 minutes.
According to Reactiv’s internal measurements, the new workflow reduced merchant configuration time by 80%, turning 17‑hour manual tasks into 3‑hour jobs and cutting the time to production by 33% (10 weeks versus 15 weeks). Job execution time fell from over 10 minutes to about 5 minutes, and the team saved roughly $6,000 per year in compute costs. The solution runs each merchant’s agents in isolated Firecracker microVMs, provides persistent cross‑session memory, and hosts the configuration‑MCP natively on AgentCore, eliminating the need for custom OpenAPI specs and authentication layers.
The story so far
4 episodes →- Reactiv cuts merchant configuration time 80% with Amazon Bedrock AgentCorethis story
How Reactiv automates mobile commerce 80% faster with Amazon Bedrock AgentCore
AWS Machine Learning Blog · 22 September 2026
Loading the full article…
This text was published by AWS Machine Learning Blog and written by Adam Gibicar. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Marketing & Small Business
All →- Higgsfield AI adds video ad features using GPT‑6 Astra · 2 src
- Opinion: use chatgpt’s three modes in order to avoid usage limits · 2 src
- How a lazy office worker uses Claude to sort tasks, draft messages, and write reports · 2 src
- Opinion: personal experience, voice, and judgment curb AI-like feel in ChatGPT articles · 5 src
- Adobe launches Premiere mobile for Android, free video editing app · 1 src
Comments
via GitHub Discussions