You can grab a Logitech wireless keyboard and mouse for $42, but the battery life is the real win

Amazon’s Prime Air is taking off in nearly 500 U.S. cities

Amazon is significantly expanding its Prime Air drone delivery service, with plans to reach nearly 500 U.S. cities by the end of 2026.

VTEX omnichannel: ship-from-store, pickup & endless aisle

A deep-dive into VTEX omnichannel configuration covering Ship-from-Store, click-and-collect pickup points, and Endless Aisle. The core prerequisite for all three modes is unified, real-time inventory visibility across every fulfillment node. Each mode requires a warehouse-loading dock-shipping policy triangle in VTEX Admin; missing any leg silently breaks checkout. Key edge cases include the hard 50km pickup radius enforced by the platform, loading dock SLA mismatches between stores and central DCs, and Endless Aisle stores lacking home-delivery shipping policies. A platform comparison table covers VTEX vs Salesforce Commerce Cloud vs commercetools on native OMS depth, implementation time, and composability. A pre-go-live checklist with owners and effort estimates is included.

Meet Penny, Pick n Pay’s new AI shopping companion

Pick n Pay is launching Penny, an AI shopping assistant built on Google's Gemini models, within its asap! delivery app. Penny lets customers build grocery baskets via text, voice, or photo — for example, asking for a recipe and getting ingredient options added to the cart. It supports multilingual input, can reference order history and Smart Shopper points via APIs, and is billed as South Africa's first conversational grocery shopping experience. Penny assembles baskets but does not yet place orders autonomously. The feature rolls out from 6 July. Pick n Pay also noted a retail-media layer is coming, allowing manufacturers to bid on conversational search terms. Rival Checkers Sixty60 launched a similar AI assistant called Pixie in April, though with a different philosophy focused on predictive replenishment rather than open conversation.

[DRAFT] Markets APIs now support MarketRegionSubdivision

Shopify's Admin GraphQL API 2026-07 release candidate introduces support for MarketRegionSubdivision, enabling Markets to be configured with country-subdivision regions like states and provinces. This is part of market-driven shipping support. Developers should use the new stable country-subdivision region type and avoid the deprecated market.regions field for sub-region membership. Currently, only shipping is supported; other configurations like Discounts, Catalogs, and Theme contextualization are not yet available and will return validation errors if attempted.

POS UI extensions 2026-07 uses per-unit fixed-amount line item discounts

Shopify POS UI extensions API version 2026-07 changes how fixed-amount line item discounts work. Previously, apps could pass a total discount for the entire line item and Shopify POS would auto-convert it to a per-unit value. Starting with 2026-07, the value passed is interpreted directly as a per-unit discount — no automatic conversion occurs. For example, passing '5.00' on a quantity-2 line item now results in a $10.00 total discount, not $5.00. Apps on 2026-04 or earlier are unaffected; apps migrating to 2026-07 must update their discount calculations to pass per-unit values explicitly. Percentage discounts are not affected.

WooMedia Matrix Pro Turns Your Media Library Into a WooCommerce Catalog

WooMedia Matrix Pro is a WordPress plugin that bulk-converts media library attachments into WooCommerce products, eliminating manual one-by-one product creation. It maps media metadata to product fields, supports batch pricing and SKU configuration, and outputs native WooCommerce data. Pricing starts at $29.99 for a single-site lifetime license, up to $149.99 for unlimited sites. WP Mayor readers get 20% off the Agency Suite with code WPMAYOR20.

TCS | Pick n Pay’s Enrico Ferigolli on Penny, the AI that shops for you

Pick n Pay ra mắt trợ lý mua sắm AI Penny trên ứng dụng asap! sử dụng mô hình Google Gemini. Penny hỗ trợ người dùng tạo giỏ hàng qua giọng nói, văn bản hoặc ảnh, lên kế hoạch bữa ăn, quản lý đơn hàng và đề xuất thay thế nguyên liệu, nhưng chưa tự động đặt hàng. Enrico Ferigolli chia sẻ lý do chọn Gemini, cách Penny tích hợp với chức năng tìm kiếm và lịch sử đơn hàng của ứng dụng thay vì truy vấn trực tiếp cơ sở dữ liệu, và lộ trình hỗ trợ đa ngôn ngữ, bao gồm tiếng Nam Phi.

Là lập trình viên AI, bạn nên đọc bài này để hiểu cách triển khai mô hình ngôn ngữ lớn (LLM) như Gemini trong ứng dụng thực tế, từ việc tích hợp với hệ thống tồn kho hiện có đến tối ưu hóa giao diện người dùng cho các tính năng như lập danh sách mua sắm tự động, giúp bạn hiểu cách chuyển đổi lý thuyết thành giải pháp thực tế trong ứng dụng thương mại điện tử.