Hunter Brennick AI Systems & Advisory ↗
AI Systems Orchestration
Part II · Patterns And Practice/Chapter 08
08

Frontier Leverage Patterns

The fun edge: multi sampling, debate, critique, generator verifier loops, search, synthetic data, and spending reasoning where it matters.

4 min read1 figures
The gist
  • "More reasoning" is expensive. Spend it where it changes the answer.
  • The winning combo: cheap models route, strong models reason, code verifies, humans judge.
  • Patterns to know: best of N, generator + verifier, planner + executor, router cascades.
LEVERAGE PATTERNS BEST OF N sample 1 sample 2 sample N judge best spend samples where variance is high GENERATOR + VERIFIER Generatordrafts the work Verifiercode · tests · judge candidate feedback code execution is the cheapest honest verifier PLANNER + EXECUTOR Plannerstrong model Executorcheap and fast replan on surprise . steps ROUTER CASCADE Routercheap classifier cheap modelmost tasks strong modelhard reasoning only escalate spend reasoning where it changes the answer
FIG 08.1Four leverage patterns that feel like AI turned up to eleven.

Module 10: Frontier Leverage Patterns

This is the fun part. These are the patterns that feel like "AI turned up to 11."

Learn the patterns that feel like AI turned up to eleven:

  • multi-sampling and best-of-N
  • self-consistency
  • debate and critique
  • generator-verifier loops
  • planner-executor loops
  • tree search over actions
  • code execution as verifier
  • retrieval plus reranking
  • synthetic data generation
  • prompt and program optimization with DSPy
  • model routing and cascades
  • memory from prior runs
  • sandbox agents that learn from artifacts
  • human calibrated rubric loops

Build by picking one hard task and running it five ways:

  • one answer
  • five answers with a judge
  • planner plus executor
  • executor plus critic
  • executor plus tests

Compare the difference.

More reasoning is expensive. Spend it where it changes the answer. The best system often follows a division of labor:

  • cheap models for routing and checks
  • strong models for hard reasoning
  • code for verification
  • humans for judgment
Next chapterChapter 09 · A Real Three-Repo Setup A real three-repo setup where AI agents do actual work: one defines the workflows, one turns requests into plans, and one does the work itself.