Notes from the build.
Engineering deep-dives and design decisions behind Empryo.
The 10 best AI coding agents in 2026, ranked by someone who builds one
Ten agents ranked on what they actually cost and actually fix, measured against real provider invoices instead of vendor claims. Includes when to pick each one, and when not to pick mine (disclosure, I build #2).
8 minread →Jul 16, 2026benchmarksI benchmarked two AI agents. Then I read the bill.
pi said $6.25. Anthropic billed $9.19. Empryo said $7.06 — billed $7.08. Self-reported agent costs are fiction until someone audits them. I keep auditing.
4 minread →Jul 16, 2026benchmarksFive real bugs, two agents, and the one nobody could fix
We stopped writing benchmark fixtures and used real merged PRs from hono, zod and ky, too new for any model to have memorized. 7/10 vs 6/10, one 8-minute hang, and a bug that beat everyone.
4 minread →Jun 1, 2025engineeringWhy an AI agent should edit symbols, not strings
String-replacement editing breaks on whitespace, drifts on line numbers, and corrupts syntax. Empryo edits the AST instead.
2 minread →May 20, 2025architectureThe Genome - a live structural map of your codebase
Instead of grepping blindly, Empryo loads a ranked map of every file, symbol, and dependency edge - rebuilt after every edit.
1 minread →