BLOG

Notes on AI-native software quality.

Practical writing from the team behind TestingOps, RunningOps, CoverOps and ReplayOps.

3 min read#AI coding agents#Code review#AI test automation

AI sped up coding and moved the bottleneck to code review: what the Harvard study says

Data from 655 firms and 720,000 workers: AI agents raised lines of code by 44% while finished work rose only 13.7%. The gap piles up in code review, where time to approval grew 56%.

Read
Cunda cover: a knowledge base of requirements, business rules, API contracts, code summaries and past defects feeding a boundary-value test case that cites its sources
4 min read#AI test automation#RAG#Test standards

AI testing is only as good as its knowledge base: RAG, test standards and summarized code

A model that writes tests knows only what you give it. Three things decide the quality of AI testing: a knowledge base fed correctly through RAG, tests that follow your test standards, and source code summarized before it reaches the model.

Read
Cunda cover: a code change with three changed lines, the two tests that reached it, a pull request gate flagging one untested line and zero-token replay
2 min read#AI test automation#Regression testing#Test coverage

AI in regression testing: run the tests your change actually touches

Regression suites grow faster than anyone can maintain them. AI helps in three places: choosing tests by the code a change touched, keeping tests working when the product changes, and keeping the cost of running them flat.

Read
Cunda cover: a virtual phone booted from OEM firmware next to a physical phone with a real SIM reading a one-time SMS code
3 min read#AI test automation#Mobile testing#DeviceOps

AI in mobile test automation: devices, one-time codes and digital twins

AI agents change how mobile tests are written, but the device and the SMS code still decide whether a flow can be tested. What agents add, where virtual devices and digital twins help, and when you need a real phone with a real SIM.

Read
Cunda cover: a ten-question checklist for comparing AI testing tools in a vendor demo, grouped as grounded, honest, cost and data
3 min read#AI test automation#AI testing tools#Agentic testing

How to compare AI testing tools: 10 questions to ask in a demo

Almost every testing tool now says it uses AI. These ten questions show the differences in a demo: where each step comes from, whether a green result can be trusted, what a run costs and where your data goes.

Read
Cunda agentic quality loop: TestingOps, DeviceOps, RunningOps, CoverOps, release gate and ReplayOps in six stages
4 min read#Agentic testing#AI#Test automation

From test automation to an agentic quality loop: how Cunda works

Agents that produce the tests, virtual phones booted from real firmware, coverage measured on running services and real user sessions. How Cunda's five products come together in one quality loop.

Read