Field notes
Notes from the work.
Deep dives on the handoffs, data leaks, and automation decisions that turn into revenue systems.
- Written for
- People running the work
- Latest posts
- 90
- Inside
- Systems, decisions, lessons
Showing 65-72 of 90 notes
Page 9 of 12
Contrarian
May 8, 2026
Why Power Automate Desktop hurts facilities ops at scale
Power Automate Desktop looks like a free win for facilities maintenance teams. Single-machine it works fine. At ten sites and real ticket volume it cracks in predictable ways.
Read the note →
Contrarian
May 8, 2026
Process automation for law firms that actually works
Most process automation pitches at law firms are RPA dressed up in legal-tech clothing. Here is what actually pays back in a 5 to 30 attorney shop, what burns money, and the operator pattern that beats the consultant model.
Read the note →
Depth Charge
May 5, 2026
Automate Work Orders From Email Without Replacing Your CMMS
Facilities teams do not need another inbox rule. They need a workflow that reads maintenance emails, extracts the work order fields, routes the request, and keeps humans in the loop for exceptions.
Read the note →
Framework + Case
May 5, 2026
Before you replace your CMMS, automate these 5 handoffs
A CMMS replacement is expensive, slow, and often unnecessary. If the pain lives in intake, dispatch, inspections, reporting, or vendor follow-up, automation around the system usually pays back first.
Read the note →
Depth Charge
May 5, 2026
Vendor dispatch and SLA tracking for facilities teams
Vendor dispatch breaks when status lives in email threads and SLA risk is noticed too late. The fix is an automation layer that sends, watches, escalates, and summarizes vendor work without replacing the CMMS.
Read the note →
Belief Shift
Apr 27, 2026
Agentic coworker pattern: role, memory, standup
An AI coworker is not a chatbot that answers questions. It is an agent with a clear role, persistent memory of its own work, and a real slot in the team's cadence. Most teams build the first and skip the last two.
Read the note →
Framework + Case
Apr 26, 2026
Anthropic SDK vs LangChain for building Claude agents
LangChain is the default. LangGraph is the upgrade path. Both add abstraction you pay for in production. Here is when the Anthropic SDK is the shorter build, the cheaper runtime, and the faster debug loop for a Claude-only agent.
Read the note →
Framework + Case
Apr 26, 2026
Claude skills vs tools vs MCP: which to reach for
Claude ships three overlapping ways to extend what an agent can do: skills, custom tools, and MCP servers. They solve different problems and most teams pick the wrong one first.
Read the note →