StackInsight
Empirical research on code performance and quality for JavaScript and TypeScript developers. Each study combines large-scale repository analysis with controlled benchmarks to produce verified, reproducible findings.
Scan your repository with Code Evolution Lab →
Research at a Glance
| Empirical studies published | 11 |
| Detection rules derived from research | 16 |
| Repositories scanned (memory leak study) | 500 |
| Prevalence of missing-cleanup patterns | 86% of repositories |
| Total findings (memory leak study) | 55,864 across 714,217 files |
| Nested loop speedup vs. Map lookup at n = 10,000 | 64× |
Code Evolution Lab
A static analysis tool for JavaScript and TypeScript codebases, built directly from the findings of this research. Code Evolution Lab detects N+1 queries, memory leaks, missing database indexes, inefficient loops, and 12 additional anti-patterns using Babel AST and Prisma schema analysis.
- App → codeevolutionlab.com
- Documentation → docs.codeevolutionlab.com
- GitHub → liangk/code-evolution-lab
Empirical Studies
Each study is structured in two phases: a corpus scan measuring the prevalence of a target pattern across real open-source repositories, followed by a controlled benchmark quantifying the performance cost of that pattern under varying load conditions. All methodology, raw data, and benchmark source code are published alongside each article.