Skip to content

From the blog

Elements & Ideas.

Product thinking, engineering craft, and observations from building at the frontier of AI.

GitHub CLI Event Streams Change How Agents Wait for Pull Requests
Testing & QA

GitHub CLI Event Streams Change How Agents Wait for Pull Requests

A practical guide to interpreting pull-request event streams without confusing one check with release readiness.

Dwayne SamuelsDwayne Samuels·
How to Test an A2UI Reservation Before It Books
Testing & QA

How to Test an A2UI Reservation Before It Books

Test stale availability, confirmation, retries, permissions and booking readback in an A2UI reservation flow.

Dwayne SamuelsDwayne Samuels·
GitHub Actions Secret Scanning for Test Logs
Testing & QA

GitHub Actions Secret Scanning for Test Logs

Learn what GitHub Actions log secret scanning covers and how to handle sensitive test logs, traces, screenshots, and recordings.

Dwayne SamuelsDwayne Samuels·
What WebHarbor Can and Cannot Prove About Browser Agent Environments
AI Agents

What WebHarbor Can and Cannot Prove About Browser Agent Environments

Learn what WebHarbor’s reconstructed websites can reveal about browser agents, where verifier and fidelity limits matter, and when to test separately.

Dwayne SamuelsDwayne Samuels·
How to Check an AI Judge Before Trusting Its Test Review
Testing & QA

How to Check an AI Judge Before Trusting Its Test Review

Learn to check an AI judge against reviewed examples and spot false approvals before relying on its test review.

Dwayne SamuelsDwayne Samuels·
GitHub’s Shared Copilot Automations Change Who Owns a QA Workflow
AI Agents

GitHub’s Shared Copilot Automations Change Who Owns a QA Workflow

A practical review card for shared Copilot Automations covering QA ownership, permissions, costs and evidence.

Samelogic TeamSamelogic Team·
What 114,134 Pull Requests Reveal About Coding Agents and Tests
AI Agents

What 114,134 Pull Requests Reveal About Coding Agents and Tests

A study of 114,134 pull requests shows coding agents add tests more often but maintain existing tests less often. Here is what reviewers should check.

Dwayne SamuelsDwayne Samuels·
How to Evaluate GPT-6.1 Terra for Visual QA
AI Agents

How to Evaluate GPT-6.1 Terra for Visual QA

Evaluate GPT-6.1 Terra for visual QA with matched tasks for perception, grounding, actions, policy, and durable outcomes.

Dwayne SamuelsDwayne Samuels·
← Prev1234567891011121314Next →

Capture browser proof before the handoff gets vague.

Select the exact element, record the replay, and give QA, product, and engineering a test artifact they can act on without another clarification loop.

Install the Chrome Extension
Visual
Semantic
Behavioral

Used by teams at

  • abbott logo
  • accenture logo
  • aaaauto logo
  • abenson logo
  • bbva logo
  • bosch logo
  • brex logo
  • cat logo
  • carestack logo
  • cisco logo
  • cmacgm logo
  • disney logo
  • equipifi logo
  • formlabs logo
  • heap logo
  • honda logo
  • microsoft logo
  • procterandgamble logo
  • repsol logo
  • s&p logo
  • saintgobain logo
  • scaleai logo
  • scotiabank logo
  • shopify logo
  • toptal logo
  • zoominfo logo
  • zurichinsurance logo
  • geely logo