Seroter's reading list: context inertia, agent evals past the vibe check, 3x code shipping
rseroter · x · 2026-10-17
Richard Seroter's daily reading list (#890) curates several substantive AI engineering pieces:
- Context inertia: agents/LLMs doggedly sticking to an initial path; the piece names the problem and suggests mitigations, including monitoring and stepping in manually.
- Agent evals: move past the vibe check—invest in real eval infrastructure, whether built or bought.
- Engineering trends: teams ship 3x more code but review it the same way; how AI reshaped Seroter's engineering leadership role.
- Also: multi-conversation Antigravity workspace setups, Reflection's Beam model getting 5x faster inference on Google Cloud via llm-d and NVIDIA Dynamo, and interactive refactoring in gopls.
More from coding & agent
- One prompt drove Opus 5.5 for ~30 hours to build a working computer from 277k logic gates — FinanceYF5 · 2026-10-17
- Grep Launches Upgraded MCP to Build and Run Agents from Claude, ChatGPT and Codex — garrytan · 2026-10-17
- Cursor's mobile app lets you babysit your AI agents from anywhere, even weekends — nateliason · 2026-10-17
- NoHumans.Space seeks testers for a knowledge network where AI agents share findings — dark-epiphany · 2026-10-17
- fx v0.0.14 ships durable agents: same API survives crashes on laptop, cloud via Vercel — cramforce · 2026-10-17
- Jeffrey Emanuel ships agent skill turning audio into spectrogram karaoke videos — doodlestein · 2026-10-17