KARIMKHALIL
I build backend services and the test infrastructure that keeps them honest — measured by what they do in production.
About
UBC Computer Science, 2026. Three internships in, most of my work has been on the systems behind the product: an order-validation API in Spring Boot at RBC Global Asset Management, the test infrastructure around it, and most recently at Minga, halving the runtime of a 3,600-test Playwright suite.
That taught me to default to evidence: tests, benchmarks, reproducible results. I like problems where correctness is measurable and the fix shows up in a number — a test run from 110 minutes to 55, a regression cycle cut by 90%, or a deep-learning pipeline's runtime by 37%.
Selected work
Big Picture
AI Portfolio & Code Analysis Platform
🏆 1st Place Capstone 🏆 Best UI/UX 🏆 Most Innovative (Students' Choice)
Live from GitHub
Every language across my public repositories, measured by bytes committed and read straight from the API. Drag to orbit the chart.
$ hover a column…
Collaboration graph · live
Every contributor I've shipped alongside, pulled live from the GitHub API and clustered by project. Node size is commit volume.Drag to orbit.
$ hover a node…
Experience
01 / 05
Software Engineer in Test | May 2026 – Sep 2026
• Led an optimization of a ~3,600-test Playwright suite, replacing per-test UI login (~8s) with role-scoped session reuse via storageState and IndexedDB, cutting full-run wall-clock time 50% (from 110 min to 55 min) and saving ~5 hrs of CI machine time per run
• Designed and implemented end-to-end Playwright test cases for product features and split monolithic spec files into smaller parallel suites, partnering with feature engineers to expand automated regression coverage
Contact