Creuto: AI Technology & Product Insights
Stay informed with practical insights from the Creuto team on artificial intelligence, custom software, mobile applications, SaaS development, product engineering, technology trends, and digital transformation.
Latest posts
More posts
JSON-LD script injection: the escape most sites skip
If structured data contains </script>, the HTML parser ends the script block early — breaking rich results or enabling script injection. The fix is one replace call, plus rendering JSON-LD where crawlers can see it.
Akash Mohapatra
Sep 17, 2026
Android developer verification: the 30 September cutoff
From 30 September, certified Android phones in Brazil, Indonesia, Singapore and Thailand stop installing apps from unverified developers, across Google Play and major manufacturer stores. Most Play apps are covered; apps distributed outside Play are the exposure.
Akash Mohapatra
Sep 17, 2026
iOS 27 App Store requirements: age ratings and SDK deadline
From September 2026, every App Store submission needs answers to new questions about social media capabilities, and from April 2027 iOS apps must be built with the iOS 27 SDK. Both can stall a routine release.
Akash Mohapatra
Sep 17, 2026
AI code review automation: how Duolingo auto-approves 10%
Duolingo lets an AI risk classifier approve about one in ten pull requests, cutting median merge time from 18 hours to 12 without a rise in defects. Hard exclusions and a literacy programme did most of the work.
Akash Mohapatra
Sep 17, 2026
EN 16931 validation: a discount has a VAT rate
Under EN 16931, a document-level discount must declare its own VAT category and rate. Invoices that treat it as a simple deduction can pass every totals rule and still misstate VAT — and clearance systems now reject them.
Nihar Ranjan Rout
Sep 17, 2026
OpenTelemetry migration at 100,000 hosts: Atlassian's approach
Atlassian replaced gostatsd with OpenTelemetry Collectors across roughly 100,000 hosts by keeping the interface services used and rebuilding everything behind it. The fixes it needed transfer to far smaller estates.
Akash Mohapatra
Sep 17, 2026
AI agent sandbox escape: VMs failed, Firecracker held
A cyber-capable model broke out of a QEMU/KVM virtual machine three times, finishing with a chain of zero-days found on its own. Against Firecracker it could crash the host but not escape.
Akash Mohapatra
Sep 17, 2026
AI agent state between turns: what does not carry over
Many failures that look like the model forgetting are environment failures. Separating model context, process state and the filesystem makes them diagnosable in about a minute.
Nihar Ranjan Rout
Sep 14, 2026
LLM as a judge evaluation: three biases to correct
Model grading reaches about 85% agreement with human annotators, above the 81% humans manage with each other. It also favours the first answer, the longer answer, and its own style.
Akash Mohapatra
Sep 14, 2026
One Vision. One Team. One Creuto.
Whatever your challenge, we bring the expertise to build, scale, and transform your digital presence with technology that works.