[{"data":1,"prerenderedAt":101},["ShallowReactive",2],{"case-study:\u002Fcase-studies\u002Fcomputer-vision-test-automation":3},{"id":4,"title":5,"body":6,"date":90,"description":91,"extension":92,"meta":93,"navigation":94,"path":95,"provenance":96,"sensitivityReviewed":90,"seo":97,"sources":98,"stem":99,"__hash__":100},"caseStudies\u002Fcase-studies\u002Fcomputer-vision-test-automation.md","Computer-Vision Test Automation",{"type":7,"value":8,"toc":79},"minimark",[9,14,18,22,25,29,32,36,39,43,46,50,72,76],[10,11,13],"h2",{"id":12},"problem","Problem",[15,16,17],"p",{},"Game and localization testing relied on people repeatedly inspecting on-screen\nvisual states by hand. That work was slow, repetitive, and hard to run\nconsistently at scale, which made it a strong candidate for automation.",[10,19,21],{"id":20},"constraints","Constraints",[15,23,24],{},"The automation had to reproduce judgments a human tester would make about\non-screen state, work across game and localization testing workflows, and fit\ninto QA processes that several teams already depended on. It also had to be\nrepeatable enough to trust in regression runs rather than one-off checks.",[10,26,28],{"id":27},"architecture","Architecture",[15,30,31],{},"I created Python automation tools for repeatable game and localization testing\nworkflows, using OpenCV to detect and evaluate on-screen visual states that\npreviously required manual inspection. Working with QA, localization,\nengineering, and project-management teams, I identified the highest-value\nopportunities to automate and led the technical direction for those\ncross-functional projects.",[10,33,35],{"id":34},"tradeoffs","Tradeoffs",[15,37,38],{},"Validating behavior through computer vision trades a degree of robustness for\nreach. Visual detection can be sensitive to changes in rendering or layout, so\nthe automation needs maintenance as the games evolve. In return, it can check\nstates that would otherwise require a person watching the screen. I focused the\nautomation on the high-repetition, high-value checks where that trade clearly\npaid off, and left genuinely judgment-dependent scenarios to human testers.",[10,40,42],{"id":41},"testing","Testing",[15,44,45],{},"The value of the automation depended on it being repeatable and trustworthy, so\nI validated the vision-based checks against the states a human tester would\nconfirm by hand before relying on them in regression runs. Increasing the\nrepeatability of regression testing was itself part of the goal.",[10,47,49],{"id":48},"results","Results",[51,52,53,62,67],"ul",{},[54,55,56,57,61],"li",{},"Reduced manual QA effort and operating costs by approximately $161k per quarter ",[58,59,60],"span",{},"docs\u002Fprofile\u002Fprojects.md",".",[54,63,64,65,61],{},"Increased the repeatability of regression testing ",[58,66,60],{},[54,68,69,70,61],{},"Allowed QA teams to focus more time on scenarios requiring human judgment ",[58,71,60],{},[10,73,75],{"id":74},"what-id-change-and-what-i-learned","What I'd Change and What I Learned",[15,77,78],{},"Vision-based automation earns its keep on repetitive, high-volume checks, but it\ncarries a maintenance cost as interfaces change. What I would carry forward is to\ninvest early in making the visual checks resilient and easy to update, and to\nkeep choosing automation targets by value rather than by what is merely\nautomatable. Leading these projects across QA, localization, and engineering,\nand over time managing teams of up to 40 people, also taught me that technical\ndirection only lands when it is chosen together with the people who run the tests\nevery day.",{"title":80,"searchDepth":81,"depth":81,"links":82},"",2,[83,84,85,86,87,88,89],{"id":12,"depth":81,"text":13},{"id":20,"depth":81,"text":21},{"id":27,"depth":81,"text":28},{"id":34,"depth":81,"text":35},{"id":41,"depth":81,"text":42},{"id":48,"depth":81,"text":49},{"id":74,"depth":81,"text":75},"2026-07-22","Python and OpenCV automation that validated game behavior with computer vision, replacing repetitive manual QA work.","md",{},true,"\u002Fcase-studies\u002Fcomputer-vision-test-automation","professional",{"title":5,"description":91},[60],"case-studies\u002Fcomputer-vision-test-automation","HodawqluyQfAklF8VNFflhfrdrf8cKsOs7RI1Csv8uM",1788403975817]