a working thesis

there is alwaysan easierway to do things.

that mindset is what brought me here.

that obsession has been with me longer than ai has.

i work with companies to find their optimal — diagnosing what's broken, building what needs to change, and measuring whether it actually did.

the gap between wanting ai and being ready for it is where the work actually happens.

§ 01 the method

the sequence.

orderautomateaugment

there is no shortcut to optimal.

each step builds the next. skip one and you are not saving time — you are borrowing against a problem you will have to undo later.

01 · order
fix the foundation.
clean the data. document the processes. find the owner of each workflow. make everything ai-friendly before the first model is opened.
this is where most companies discover they have more work than they expected.
02 · automate
remove the friction.
clean triggers. logged outcomes. deterministic systems that run without supervision.
most problems stop here. that is not failure — that is the optimal solution.
03 · augment
add intelligence where it earns its place.
not where it creates novelty. not where it impresses in a demo. where it changes the outcome on a regular tuesday.

most teams skip the first two steps and wonder why ai does not survive past the second quarter.

§ 02 readiness

you are ready when
five layers hold.

a five-layer audit before any ai conversation. each layer has to hold before the one above matters.

01 · data
is it captured, clean, and reachable?
02 · process
is the workflow documented and stable?
03 · ownership
who owns the outcome and the failure mode?
04 · measurement
is success defined in numbers, not vibes?
05 · adoption
will the team actually use it on monday?
see the full framework →
§ 03 the notebook

short essays.

published when the thinking is finished — not when the calendar says so. each note pins one idea worth keeping.

008
why most companies are not ready for ai yet
the problem is almost never the model. it is what sits underneath it.
ai readiness
007
automation before intelligence
half of the ai projects i have audited would have shipped more value as a clean script.
practical implementation
006
ai adoption is mostly a human problem draft
the technology is rarely the bottleneck. the calendar, the trust, and the quick win are.
adoption
005
the operational bottlenecks behind failed ai projects draft
three patterns repeat across every post-mortem i have read.
workflow intelligence
004
when not to use ai draft
a one-page decision rule before adding a model to anything.
practical implementation
all notes →
§ end closing

most companies want ai.

most are not ready for it yet.
that gap is where the real work is.

the search for optimal never ends — and it rarely starts where you think it will.

if you are trying to figure out where your company actually stands, what needs to be fixed before ai makes sense, or simply how to make things work better before reaching for the next tool — that is the conversation i find most interesting.

say hello → read the manifesto →