Skip to content

Notes

Field notes

Short, practical writing on building AI-native systems. Filter by tag.

AI search is a citation problem
Industry6 min read

AI search is a citation problem

AI Overviews and AI Mode changed how buyers research. The sites that win are the ones an answer engine can safely cite. Here is the checklist.

Voice agents are becoming front-desk systems
Industry6 min read

Voice agents are becoming front-desk systems

Real-time multimodal models are moving voice agents beyond scripts. The business case is intake, routing, and follow-up with clear consent.

Measure AI referrals like a channel
Industry6 min read

Measure AI referrals like a channel

45% of consumers now ask AI tools for local business recommendations. How to see that traffic in analytics, your lead records, and intake questions.

Write the one-page AI policy
Operations5 min read

Write the one-page AI policy

Most small businesses using AI have no written rules. A one-page policy that covers tools, data, approvals, and receipts without slowing anyone down.

Plan for the model going away
Operations5 min read

Plan for the model going away

Claude Fable 5 was offline from June 12 to July 1. What model outages, deprecations, and price changes mean for workflows pinned to one model.

Computer-use agents need production wrappers
Cutting Edge7 min read

Computer-use agents need production wrappers

Browser and shell agents are ready for bounded work when scopes, screenshots, approval points, budgets, and receipts are built into the wrapper.

MCP needs an operating model
Operations7 min read

MCP needs an operating model

MCP servers need inventory, scoped credentials, consent rules, logs, and receipts because tool access becomes a business surface fast.

Agents only pay when workflows change
Fundamentals6 min read

Agents only pay when workflows change

AI adoption pays when a team changes the workflow: pick one clean input, define the output, measure accepted work, and keep review in the loop.

Model vs harness vs environment
Fundamentals5 min read

Model vs harness vs environment

A debugging frame for agent failures: separate model judgment from harness rules and environment issues before paying for a stronger route.

Why we built our own agent harness
Industry6 min read

Why we built our own agent harness

A practical look at the control layer behind Om Concepts agents: client sandboxes, tool rules, budget gates, receipts, and where vendor tools still fit.

Receipts prove each run
Operations5 min read

Receipts prove each run

Agent receipts show model routes, prompt hashes, tool calls, token costs, review state, and what clients can inspect without exposing private work.

Service pages that help buyers decide
Operations5 min read

Service pages that help buyers decide

A service-page structure for showing who the offer fits, what gets shipped, what shapes price, which proof matters, and what to do next.

Where small businesses lose leads
Operations5 min read

Where small businesses lose leads

A lead handoff map for small teams: match ad promises, capture useful fields, set reply ownership, and automate only after the loop works.

Local SEO operating cadence
Industry6 min read

Local SEO operating cadence

A 90-day local SEO cadence for listings, reviews, useful content, technical hygiene, and tracking whether visibility becomes pipeline.