Agents made implementation cheap, so I started doing far more of it in parallel. My GitHub history since 2019 shows what that did: as VP of Technology I am back at my full-stack engineer pace, with bursts, burnout and a low season that is no longer low.
Agents write all my code now. I still keep asking them to simplify it, because unnecessary complexity creates more to reason about and more places for bugs.
How OpenCode's database, TokenScope, and a fresh greeting helped me trace our agent's token usage to MCP tool definitions—and what Code Mode changed.
How I used coding agents, temporary compatibility layers and a lot of review to complete an oRPC contract migration I would never have attempted by hand.
How forgotten Git worktrees became Omaprune, why Omarchy made it easy to share, and where the plugin trust model still needs work.
A practical retrospective on building and rolling out an internal company agent at Bold.org, from connectors and usage analytics to adoption, contributions and overbuilt evals.
How my engineering workflow moved from coding alongside one assistant to supervising parallel agents, and why company context became the real breakthrough.