Skip to content

About the Playbook

Read enough teardowns and the same hard problems surface again and again, solved in recognizably similar ways. The Applied AI Playbook lifts those problems out of any single company and treats each one on its own: what makes it hard, the patterns that recur, the tools and popular choices teams reach for, a reference architecture, and the best practices — each tied back to the teardowns it’s drawn from.

This is the cross-company comparative map the teardowns point toward. It’s synthesis, not a per-company reconstruction: the unit here is the problem and its solution space, evidenced by linking to the teardowns where each move shows up — not a per-claim confidence tier (that discipline lives in the teardowns themselves). For the vocabulary underneath these patterns, see AI Engineering Terms.

Each row is a problem common to applied-AI products and the page that works it through. Pages link to the teardowns that supply the evidence.

ChallengeThe problem in one lineSeen across
Graduating an agent from assistant to actorCrossing from “suggest” to “act” without losing user trustAntimetal, Prophet Security, Pallet, Basis, Confido, Amperos
Testing output that isn’t reproducibleNo fixed expected output, so a normal test suite can’t gate changesGlean, Rilla, Traba, Momentic, Basis
Keeping inference cheap & fastFrontier-model calls on every step blow up cost and latencyBasis, Glean, Momentic, Traba
Reaching systems with no clean APIThe systems of record are legacy portals built for humans, not machinesPallet, Amperos, Confido, Momentic
Retrieval at multi-tenant scaleGrounding every tenant in its own knowledge without per-customer codePallet, Glean, Rilla
Surviving long, multi-day workflowsMulti-party flows must survive crashes, retries, and partial failurePylon, Gradient Labs, Harvey
Beating context degradationOne agent’s context grows until quality quietly falls offTraba, Antimetal, Basis
Encoding dense domain rulesRegulatory and underwriting logic must be exact, testable, and auditablePylon, Basis, Comp AI
Own vs. rent the modelDeciding where to spend a training budget vs. renting a frontier LLMRilla, Basis, Harvey, Pallet, Glean