Article · 2026-07-03 · 6 moments

Fable's judgement

One of the most interesting tips I got from the Fireside Chat I hosted with Cat Wu and Thariq Shihipar from the Claude Code team at AIE on Wednesday was to let Fable (and to a certain extent Opus) use their own judgement rather than dictating how they should work. The example they gave was testing. You can tell Fable "only use automated testing for larger features, don't update and run tests for small copy or design changes" - but it's better to just tell Fable to use its own judgement when dec ✦ AI generated

Read original ↗

Claim◆

Fable, and to a certain extent Opus, work best when allowed to use their own judgement about how to do a task rather than being given dictated, rigid instructions.

A tip from the Claude Code team: let Fable and Opus decide how to approach a task themselves instead of prescribing exact rules.

transcript

Cat Wu and Thariq Shihipar: One of the most interesting tips I got from the Fireside Chat I hosted with Cat Wu and Thariq Shihipar from the Claude Code team at AIE on Wednesday was to let Fable (and to a certain extent Opus) use their own judgement rather than dictating how they should work.

gives example · 3

Anecdote

For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent.

Simon put the advice into practice by prompting Claude Code to judge which lower-power model to delegate each coding task to via a subagent.

transcript

Simon Willison: I prompted Claude Code just now with: For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent Claude saved this memory file in ~/.claude/projects/name-of-project/memory/delegate-coding-to-subagents.md:

explains mechanism · 1extends · 1supports · 1

Mechanism

Implementation work rarely needs the top-tier model, so coding tasks should be spawned as subagents with a model override (Sonnet for substantive implementation, Haiku for trivial/mechanical edits), while judgment, review, and synthesis stay with the main loop.

The saved memory file spells out the rationale and rule: hand off routine coding to cheaper subagent models, but keep judgment-heavy work (design, review, synthesis) in the main model.

transcript

Claude: Why: cost/efficiency — implementation work rarely needs the top-tier model; judgment, review, and synthesis stay with the main loop. How to apply: when a task in this project is primarily writing/editing code, spawn an Agent with a model override (sonnet for substantive implementation, haiku for trivial/mechanical edits) and a self-contained prompt; review the result in the main loop before committing.

Example

Rather than giving Fable an explicit rule like 'only use automated testing for larger features, don't update and run tests for small copy or design changes,' it's better to tell Fable to use its own judgement about when to write tests.

Testing was the concrete example: instead of hard-coding a rule about when to test, just tell Fable to judge for itself.

transcript

Cat Wu and Thariq Shihipar: The example they gave was testing. You can tell Fable "only use automated testing for larger features, don't update and run tests for small copy or design changes" - but it's better to just tell Fable to use its own judgement when deciding to write tests instead.

Anecdote

This delegation approach is working well — Simon is getting a lot done and his Fable allowance is shrinking less quickly than before.

Since adopting the judgement-based model-delegation approach, Simon reports strong productivity with slower depletion of his Fable token allowance.

transcript

Simon Willison: Design, auditing, data synthesis, and anything judgment-heavy stays in the main model. See also [[project-goals]]. So far it seems to be working well. I'm getting a ton of work done and my Fable allowance is shrinking less quickly than before.

Claim◆

To avoid burning through valuable Fable tokens before the price increase, tell Fable to hand off smaller tasks to other models, letting it use its own judgement about which model to use.

Jesse Vincent's tip: have Fable judge when a task is small enough to route to a cheaper model, saving tokens ahead of an upcoming price rise.

transcript

Jesse Vincent: Jesse Vincent just gave me a related tip to help avoid burning too many of those valuable Fable tokens in the few days we have left before the prices go up. Tell Fable to use other models for smaller tasks, applying its own judgement about which model to use.

Highlight slides

Let the Model Decide How✦ from: Fable, and to a certain extent Opus, work best when allowed to use their own judgement about how to do a task rather than being given dictated, rigid instructions.Save Fable Tokens Before Price Hike✦ from: To avoid burning through valuable Fable tokens before the price increase, tell Fable to hand off smaller tasks to other models, letting it use its own judgement about which model to use.Where the Tip Came From✦ from: Fable, and to a certain extent Opus, work best when allowed to use their own judgement about how to do a task rather than being given dictated, rigid instructions.How the Handoff Works✦ from: To avoid burning through valuable Fable tokens before the price increase, tell Fable to hand off smaller tasks to other models, letting it use its own judgement about which model to use.

Related episodes

Article · The Pragmatic Engineer

Fable, and to a certain extent Opus, work best when allowed to use their own judgement about how to do a task rather than being given dictated, rigid instructions.

For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent.

Implementation work rarely needs the top-tier model, so coding tasks should be spawned as subagents with a model override (Sonnet for substantive implementation, Haiku for trivial/mechanical edits), while judgment, review, and synthesis stay with the main loop.

Rather than giving Fable an explicit rule like 'only use automated testing for larger features, don't update and run tests for small copy or design changes,' it's better to tell Fable to use its own judgement about when to write tests.

This delegation approach is working well — Simon is getting a lot done and his Fable allowance is shrinking less quickly than before.

To avoid burning through valuable Fable tokens before the price increase, tell Fable to hand off smaller tasks to other models, letting it use its own judgement about which model to use.

Slow down to speed up: so much has changed in 6 months’ time

Import AI 464: Fables writes GPU kernels; AI automation; and analog computation

[AINews] Claude Tag: Multiplayer, Proactive, Persistent Agents in Slack