https://blog.blackwell-systems.com/tags/precision/Recent content in Precision on Blackwell SystemsHugoen-usSun, 24 May 2026 00:00:00 +0000https://blog.blackwell-systems.com/posts/ai-code-context-tools-benchmark/Sun, 24 May 2026 00:00:00 +0000https://blog.blackwell-systems.com/posts/ai-code-context-tools-benchmark/codegraph has 19K GitHub stars. Aider has 20K. We benchmarked 7 systems on 117 tasks across 7 codebases (3.5M LOC to 14K LOC). knowing is 1.36x more precise than codegraph, 2.45x vs GitNexus, 2.92x vs Gortex, 14.2x vs grep. Queries Kubernetes in 2ms (codegraph: ~1s), eliminates 99.9% of grep noise on ambiguous queries.https://blog.blackwell-systems.com/posts/ai-agent-code-search-precision-benchmark/Fri, 22 May 2026 00:00:00 +0000https://blog.blackwell-systems.com/posts/ai-agent-code-search-precision-benchmark/Rigorous benchmark of AI agent code retrieval: 107 tasks, 5 repos, 5 languages, 4 competitors. grep precision: 2%. GitNexus: 7.6%. knowing: 23% (11.5x better, p<0.0001). Plus: 193x faster indexing, 28x less RAM, 48x more token-efficient than Repomix. The first statistically validated comparison of code intelligence tools for AI agents.