AI models and agents on the business line
Model routing
Model choice and routing
- How to Cap the Strong Model in a Per-Task Router
- How to Choose a Model for Ticket Triage on Your Own Cases
- How to Choose an AI Model for Your CRM Notes
- How to Choose the Number of Score Bands for a Tiered Workflow
- How to Choose the Second Model When Its Only Job Is Reviewing
- How to Route Bank Rows Between a Cheap and a Strong Model
- How to Route Between a Cheap and an Expensive Model
- How to Route Contract Review Between a Cheap and a Strong Model
- All of Model routing
Prompt design
Prompt design and guardrails
- How to Brief a Model on Your Public Facts So It Never Invents One
- How to Prompt a Model for Disconfirming Evidence
- How to Set Chunk Boundaries So a Fact Survives Whole
- How to Size a Chunk for a Table So a Row Stays with Its Header
- How to Stop a Reviewer from Rewriting the Entry
- How to Write a Decision Rule an Agent Can Follow
- When the Fact Sheet Becomes Retrieved Context: A Sizing Rule
- How to Batch Many Transcripts per Call Without Losing the Quote
- All of Prompt design
Evals and gold sets
Evals and gold sets
- How Many Evals Are Enough? A Sample Size Rule for Model Testing
- How to Build an Eval Set from Your Support Tickets
- How to Compare Two Codebook Versions on the Same Batch
- How to Eval Whether Your Pipeline Honors Exclusions
- How to Set a Minimum Accuracy per Ticket Category
- How to Test a Schema Card Against a Set of Planted Questions
- How to Test Whether a Disclosure Reviewer Catches a Planted Leak
- How to Turn Adherence Scores into a Batch Go/No-Go Rule
- All of Evals and gold sets
Human review
Human review and escalation
- How to Brief a Second Model to Review the Same Output
- How to Design a Junior-Clearable Agent Output Review Queue
- How to Train a Second Reader on Your Go/No-Go Rule
- How to Write a Reviewer Prompt That Abstains
- How to Write a Second-Model Reviewer Prompt from the Same Rubric
- How to Write an Escalation Rule for Model-Human Disagreement
- When to Give the Model the Final Say on a Checkpoint
- When to Use AI to Draft Public Updates (and When Not To)
- All of Human review
Agent permissions
Tool permissions and agent autonomy
- How to Write a Scope Boundary an Agent Build Cannot Drift Past
- How to Write a Tool Call a Rule Can Rely On
- How to Write an Idempotency Key an Agent Can Derive
- How to Cap What an Agent Can Spend or Commit Without Asking
- How to Choose a Timeout and Step Limit for an Agent Loop
- How to Design a Deduplication Store an Agent Can Query
- How to Design a Refusal Message an Agent Can Emit and Log
- How to Measure an Action's Reversibility Before Final Say
- All of Agent permissions
Agent supervision
Agent handoffs and supervision
- How to Catch an Agent Inventing a Number from the Wrong Source
- How to Catch an Interviewer Agent That Drifts Off the Guide
- How to Hand Off Work to an AI Agent Without Burning Out the Team
- How to Split a Coding Task Between a Model and a Person
- How to Write a Review Rubric That Makes Agent Supervision Faster
- How to Compare Two Interviewer Agents on the Same Guide
- How to Decide Which Queued Tasks Expire Instead of Waiting
- How to Estimate a User's Tolerance for Waiting on an Agent Call
- All of Agent supervision