Every's 'frontier team' tracks new models and audits AI spend per person and task
every · x · 2026-10-27
Every shares its "frontier team" practice: a small crew whose job is testing each new model, assessing its impact on products and workflows, and turning useful experiments into tools and articles while the rest of the company stays on the roadmap — Dan Shipper calls it "kind of transformative." The post also demos Every's usage dashboard that breaks AI spending down by person, model, and effort level, plus a companion workflow that groups sessions into work types (user-approved categories) so staff can judge whether spending was worth it. The episode also covers a code-review skill for non-coders and using AI for a cybersecurity incident.
Related event: Every's Frontier Team Tracks Every New AI Model Full-Time(3 posts)→
More from coding & agent
- Kolab Benchmark Uses Pro Screen Recordings to Show AI Agents Overestimate Collaboration Quality — minnesotanlp · 2026-10-27
- Dev argues Claude burns far more tokens than GPT models for the same coding work — haider1 · 2026-10-27
- Stop building agents that can't change models: stable contracts, external state, real evals — rseroter · 2026-10-27
- Cheap models can cost more in agent harnesses than frontier ones, researcher warns — vasuman · 2026-10-27
- Kanzo: A 5k-Line Rust Kanban Binary With MCP for Claude Code and Codex — allmudi · 2026-10-27
- 'A human approves everything' is a trap: approval fatigue and how to actually design safe agents — Efistoffeles · 2026-10-27