SERVICES

I implement test automation you can keep — Playwright, CI, and agents that run against the real suite. Not a slide deck. Not a staff-aug team. Me, in your repo, until the pipeline is boring in the best way.

01-framework.sh

Test framework implementation

Playwright from a blank repo: architecture, fixtures, tagging, and the boring bits that keep a suite honest after month six.

02-ci.sh

CI/CD testing integration

Merge queues, deployment gates, sharding, and flake isolation so engineers trust the pipeline instead of retrying it.

03-infra.sh

Quality infrastructure

Centralized reporting and metrics so five teams are not five different definitions of “done.”

04-train.sh

Team training

Workshops and pairing. The goal is that you do not need me in the room to add the next spec.

ben@spurlock:~/services/ai$

man ai

AI in the test loop

The useful version is not a chatbot bolted onto QA. It is MCP, repair agents, and failure triage wired into Playwright the same way your CI is.

05-mcp.sh

MCP for the test repo

Give agents the same tools you have: run Playwright, pull traces, read CI logs, and inspect fixtures without copy-pasting stack traces into chat.

06-faster.sh

Faster test development

Agents draft coverage from tickets and PRs so you spend time on the assertions that matter, not boilerplate page objects.

07-repair.sh

Agents that fix broken tests

When a selector or wait dies, an agent proposes a patch with the failing trace attached. You review. The suite does not sit red for a week.

08-triage.sh

Report real failures

Classify flake vs product bugs, open the issue with repro steps, and stop treating every red build like a pageable incident.

mail

ben@spurlock:~$ cat next-steps.txt

Let's discuss your testing challenges

Tell me about flake, pipeline time, or the framework nobody wants to own. I will tell you if I am the right person.

./talk.sh