Building a More Efficient Agent -- Laura Entis (Every)
Every's Context Window issue combines three strands: (1) a podcast recap of Dan Shipper's conversation with Willie Williams on why Every retired personal agent-pets in favor of one shared company agent, now shipped as the Every Agent; (2) an engineering case study on how staff engineer Paridhi Agarwal cut Every Agent token costs on 11 frequent tasks by over 80 percent, including a prompt-caching bug nobody knew they had; and (3) the issue's recurring sections -- this week's model picks across the team, seven new Thesis Statements, and a Jagged Frontier essay on AI-first engineering management.
The core case study: an early attempt to route all Every Agent tasks through a cheap coordinator model (Sonnet) that escalated complex jobs to Opus backfired -- quality and cost both got worse, because many requests ended up processed by both models. What actually worked: (a) upgrading Opus 5 to Opus 5.5, which improved response quality and cut token costs within hours of release; (b) rewriting tool responses (GitHub, Linear) to return shorter, labeled results and loading full tool descriptions only on demand (~10% reduction); and (c) discovering that Claude auto-injects each user's Slack ID into memory-folder names inside the system prompt, which silently defeated prompt-cache reuse across users with zero error signal -- "the team assumed caching was working. It wasn't." Fixing it (generic folder names, Slack IDs kept out of the visible prompt) cut costs by 74 percent in Paridhi's best-case test, stacked on the other two fixes.
Why this is in the vault
Direct, measured engineering lessons (a coordinator-routing failure mode, a specific prompt-caching bug pattern, a tool-response-size fix) for any shop running Claude-based agents in production, plus a continuation of the personal-vs-shared-agent thesis this vault is already tracking.
Mapping against Ray Data Co
Strong mapping. The most concrete transferable finding is the caching bug: Claude silently injects user-specific identifiers (Slack IDs via memory-folder names) into the system prompt, which breaks prompt-cache reuse with no visible error -- just a quietly inflated bill. That is a directly checkable risk for RDCO's own Claude-based agents, starting with the always-on channels-agent ([[project_channels_agent_setup]]): does its system prompt or memory-folder naming vary per-conversation or per-chat-id in a way that defeats caching the same way? A 70%+ silent token leak wouldn't show up in cron logs or output quality, only in the bill -- worth a direct audit rather than assuming caching "just works." Second, the coordinator-routing failure (a cheap model told to self-triage and escalate hard work made both quality and cost worse, because requests got double-processed) is a sharp cautionary data point against any naive "router decides, bigger model executes" pattern, directly relevant to how RDCO already pairs model and effort explicitly across sub-agent dispatch rather than trusting a model to judge its own limitations (feedback_delegation_model_effort_pairing). Third, Willie Williams's personal-agent rule of thumb from the podcast half -- a transactional task gets a disposable Codex-style thread, a relationship-building task gets its own named persona -- is a usable design heuristic for Ray's own mascot/persona work (feedback_ray_mascot_instantiation_pattern), and it reinforces, with a new mechanism (prompt caching + tool-response cost, not just adoption virality), the shared-vs-personal-agent thesis the 10-06 GA-launch note already banked for RDCO's agent-deployer positioning.
Curation section
- Daily Driver (model picks, this week): Opus 5.5 dominant for coding/creative work among engineers and designers; GPT-6.1 Sol for admin, scheduling, and computer-use tasks; a few early-access unreleased models in play (Dan Shipper's CEO pick, Katie Parrott's "seeeeecret model"); one explicit note that Fable 5.1 usage has stopped entirely in favor of Opus 5.5 -- "I even stopped using Fable 5.1 completely and now I'm rarely running out of usage" (Nityesh Agarwal, senior applied AI engineer).
- Thesis Statements (7 new entries): Contestable one-line claims on the future of AI and human work from Every staff and guests, including Willie Williams's "History will be rendered, not read" and Kieran Klaassen's "Nothing will go wrong twice" (Klaassen created Compound Engineering).
- Jagged Frontier essay (Willie Williams): A practical setup for AI-first engineering management -- transcribe every meeting, feed the transcripts plus GitHub PR activity to an AI digest tool (Every uses an internal tool called Tend), and use that feed to get direct organizational signal instead of relying on the usual layers of managerial retelling.
Related
- [[2026-10-06-every-agent-launch-shared-slack-coworker]]
- [[2026-05-15-every-team-agents-vs-personal-pets]]
- [[2026-09-22-every-vibe-check-opus-5-5-codex-converts]]
- [[project_channels_agent_setup]]
- [[feedback_delegation_model_effort_pairing]]
⚠️ Sponsorship
House self-promotion throughout: the piece exists to sell subscriptions bundling access to the Every Agent (pass-through token pricing, "no markup" framing repeated twice in the issue), and the footer cross-promotes Every's own product suite (Sparkle, Cora, Spiral, Monologue). The token-efficiency case study itself reads as genuine engineering detail rather than puffery, but the "cut costs 80 percent" framing is still a vendor demonstrating its own product is worth the subscription price, not an independent benchmark.