Systems | Development | Analytics | API | Testing

“Customers can apply a discount code” looks ready for testing. Give that sentence to an AI model and it can produce cases for valid, invalid, expired, and reused codes in seconds. The trouble starts when someone asks what a valid code should do. Which basket is eligible? Does the discount change the displayed total before or after tax? Should the saved order carry the same amount?

In one release test, my AI agent triaged 43 backlog items into 134 executable cases in Katalon True Platform. The first pass at the steps caused trouble. We had bundled setup, navigation, action, and judgment into long instructions. The runner returned verdicts, but a failed step left me unsure which expectation it had checked. We split the cases into one action and one checkable expected result per step. The failures became specific enough to review against the screenshots and session recording.

Part 1 ended with a question for anyone running an internal app: which sources does its most important output depend on? I thought I knew the answer for our marketing dashboard. On 9 August 2026 I found out I did not. Two months later, all 14 of our marketing sources land in one place, and the numbers we act on have changed. The full before-and-after list is in the second tab of the data map further down.

This year, AI agents have been writing more code than ever, and we've seen the slow part of shipping move from writing code to building and releasing it. You've probably felt it in your own team too. So this quarter we focused on making sure all that extra code doesn't turn into extra waiting for you. Here are the 20+ updates we shipped across Bitrise in Q3.

Combine off-the-shelf data integration and transformation with Lakeflow’s streaming data replication, orchestration, and platform-native integrations.

AI agents can write and run tests faster than most frameworks can keep up. Direct control of shared real devices allows these agents to recover when unexpected issues arise during testing.

C By Chris, Head of Content at LoadFocus·

B By Bogdan, Founder at LoadFocus·

When the news about OpenAI's agents broke on September 26, 2026, my first reaction wasn't shock. It was recognition. If you missed it: OpenAI's agents were running what should have been routine research tasks during training and evaluation, and they ended up on US government websites. They reached the SEC and the Census Bureau, and their activity touched the Justice and Commerce Departments and government sites in five states.

New addition to LeoCore offers real-time, intent-driven orchestration that replaces manual coordination at scale.