Notes from teams building
revenue-grade agentic systems.
Writing on AI architecture, operating models, and the unglamorous work of moving experiments into production. From our engineers, partners, and operators in the field.
- 42
- Articles
- 6
- Series
- 12k
- Subscribers
The agentic system is the new org chart.
When AI takes ownership of a workflow, it inherits the same operational rules as a hire: a scope, a KPI, an escalation path, and a manager. We've spent twelve months retiring the "automation" frame and replacing it with something that looks a lot more like staffing.
What we've been writing about this quarter.
Speed-to-lead is a routing problem, not a hiring one.
Why most teams chase headcount when the bottleneck is the first 90 seconds of a lead's life, and how to design an agent for that window.
A working definition of "revenue-grade" for AI systems.
Three tests we apply before we let an agent touch the P&L: ownership, escalation, and a KPI the CFO can verify on Monday.
How brands get cited (and excluded) in AI answer engines.
We tracked 600 prompts across four answer engines for sixty days. Here is the citation pattern, and the seven structural fixes that moved the needle.
Designing escalation rules an agent actually obeys.
The difference between an agent that hands off cleanly and one that buries problems is twenty lines of policy and a single observable signal.
How a $40M ops team retired 32 SOPs in six weeks.
A field report from a multi-site fulfillment operator that replaced its dispatch logic with a single decision agent — and what broke first.
The four-loop marketing engine: brief, ship, measure, prune.
Most AI-marketing setups stop at "ship." We argue the prune step — killing content that isn't earning attention — is where the real lift comes from.
Long-form reading.
-
FN/035
Architecture
What "context" actually means inside a working agent.
14 APR -
FN/034
Forecasting
Why your CAC dashboard is two weeks behind your reality.
08 APR -
FN/033
Case Study
A B2B sales team that cut its qualification cycle from 11 days to 38 hours.
02 APR -
FN/032
AI Search
Schema, citations, and the small structural fixes that move answer share.
27 MAR -
FN/031
Sales Ops
Inbound is a queue. Treat your AI sales agent like a router, not a closer.
21 MAR -
FN/030
Architecture
A defensible separation of concerns for multi-agent systems.
15 MAR -
FN/029
Marketing
What "brand voice" actually means to a generation model.
09 MAR
Multi-part deep dives.
-
S/01 · 5 parts
The AI org chart
Designing agentic roles with ownership, scope, and a real KPI. Lessons from twelve production deployments.
-
S/02 · 4 parts
Revenue-grade forecasting
Building forecasting systems whose outputs make it onto the next board deck — without the asterisks.
-
S/03 · 3 parts
Visibility in AI search
How brands win citations across answer engines, and what gets you excluded from the final answer.
-
S/04 · 6 parts
The owner's chief of staff
A working build of an AI chief of staff: briefs, memory, escalation, and the ten things it should never decide.
-
S/05 · 4 parts
Marketing in four loops
Brief, ship, measure, prune. A complete operating system for content teams that aren't ready to disappear.
-
S/06 · 3 parts
Engagement field reports
What broke first, what we did about it, and what the team kept after we left. Three full deployments, no smoothing.