Blog

AI agents, RAG, and integrations — notes from production.

How to Choose Between an AI Agent and a Fixed Workflow

AI agent vs workflow: an agent decides the steps at runtime; a workflow runs steps you coded with the LLM in slots. Prefer the workflow when the path is known.

Human-in-the-Loop AI: Where to Put the Person

Human-in-the-loop AI: the person sits before the action, after it, on the exceptions, or on a sample - four placements, each with different trade-offs.

Retrieval-Augmented Generation: The Failure Modes Nobody Demos

RAG failure modes: weak retrieval grounding a confident wrong answer, stale content, chunking that splits the answer, and losing the source citation.

How to Keep an LLM Feature's Cost Predictable as You Scale

LLM cost control at scale: route simple calls to a small model, cache the head of the distribution, cap calls per task, and budget for the peak multiplier.

Prompt Versioning: Treating the Prompt as Code

Prompt versioning means the prompt is code - in version control, pinned per decision, tested against an eval set, and rolled out and back like any other change.

What a Production-Ready AI System Needs That a Demo Doesn't

AI demo vs production: the demo shows the happy path. Production needs failure handling, eval sets, audit logging, cost controls, and a human escalation route.

How to Scope an AI Project So the Timeline Doesn't Slip

AI project scoping turns a vague ask into a buildable spec: non-functional constraints, one hard requirement, a first-release cut, and what's out of scope.

Build vs Buy for AI: A Framework That Works Across Fraud, Search, and Support

Build vs buy for AI is three options: buy a product, build from scratch, or a custom build on managed components. The signals that move you between them.

How Much Does a Custom AI Feature Cost to Run Per Month?

AI feature run cost has three parts in every domain: model inference at your volume, supporting infrastructure, and the human review that never goes to zero.

How to Run an AI Pilot That Produces a Real Go/No-Go Number

How to run an AI pilot: one narrow slice, a go/no-go metric and threshold set before you start, a control to measure against, and the run cost too.