Volume finally matters: how agents are rewriting software engineering productivity
latticecut · x · 2026-10-09
- poteto (ex-Shopify) argues that agents have inverted a long-held engineering norm: lines of code and PR counts never mattered because impact did — one config change could save millions.
- Pre-agents, volume only surfaced in negative contexts, like low performers padding PRs; the metric was rightly distrusted.
- With agents cheaply generating code at scale, an engineer's real job shifts to reviewing and gatekeeping massive agent output, so throughput and volume re-enter the productivity conversation.
- He acknowledges the skepticism is earned but says the old impact-only framework no longer holds; teams need new evaluation models for the agent era.
More from coding & agent
- Keyfleet: agent swarms with onchain treasuries airdrop first Keys to Regents holders — seanwbren · 2026-10-10
- User says Grok bot's hidden subagents and instant replies ruin other LLM experiences — rudrank · 2026-10-10
- Alma agent edits an a16z-style video in Premiere Pro fully via computer-use — itsOmSarraf_ · 2026-10-10
- Developer lets Claude work overnight via Amp Code, self-training an on-device private classifier on a Mac Mini — iannuttall · 2026-10-10
- loop-engineering hits 11.4k GitHub stars with CLI tools for orchestrating AI coding agent loops — tom_doerr · 2026-10-10
- From Claude Code to Pi: dev open-sources 'Benchmaxxed' extensions for the token-efficient harness — Remote_Book · 2026-10-10