Dinand Tinholt (@tinholt) header
Dinand Tinholt (@tinholt) profile picture

Vice President @Capgemini. | Building at the intersection of AI, data, and business transformation | Views my own

Chicago, IL
Joined February 2009
Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)Photo by Dinand Tinholt (@tinholt)
9,000 real agent decisions, three Claude models, five back-office workflows, four oversight regimes. Pre-registered. Full human approval never came out on top.
2
1
24
What happens when you send more agent work to review but keep the review team the same size? I ran 90,568 simulated configurations to find out.
1
25
Fixed team: the delegation margin falls, slope −0.593. Grow the team in step with the load: it rises, +1.148. Same routing decision, opposite sign.
1
18
90,880 configurations of agent organisations. Delegation depth weighed more against quality than anything else in the model. The first version had one flaw: its reviewers never got tired.
1
22
Let them tire and the lever drops from +1.46 to +0.93, turning negative past 80% sustained utilisation. Scaling staff with the escalations scored +1.15. Fixed headcount scored -0.59.
1
11
I rebuilt the game that got me into computers, with a model under my desk and a day of agents. A small 3-bit open-weights model rebuilt a 1988 engine from a spec: 503 tests, 78 hours, 26 kWh.
1
2
44
Then a blind gauntlet against a commercial bar built Mercy County in a day. Spec and tests first, cheap local build second. A ratio that applies far beyond games.
1
18
A fine-tuned encoder first reached its workflow ceiling at 1,000 labelled cases: 0.825 vs 0.822 on challenge cases. The from-scratch classifier never got there, even at 3,000. And the shipped eight-vote mode bought nothing. Claim withdrawn. medium.com/@tinholt/how-many…
2
62
At $11.65 in video-generation charges, a creative idea becomes something you can afford to explore. Chicago in Five Breaths: 51 seconds of watercolor Chicago, Belmont Harbor to Navy Pier after dark. dinand.com/side-projects/chi…
1
52
An architecture diagram starts a debate about the architecture. My second Maison Agent experiment presents the AI control plane like a fragrance ad: one gold ring gathering black ribbons, one sentence beneath it.
1
1
36
I built Maison Agent: a parody luxury house for AI agents. 43 specialists, 13 industries, agent portraits as sculptures, prices by consequence — $59 onboarding, $99 safety triage. (1/2)
1
1
31
The joke's serious edge: every output ends in small type — 'No message has been sent. No order has been placed.' Luxury sells what you can't inspect. An enterprise agent is the same problem, higher stakes. medium.com/@tinholt/what-a-p… (2/2)
32
How often should you read what your agents did? A pre-registered simulation: at 800 reviewer tokens per fleet-hour, hourly monitoring caught an injected fault 2.02 fleet-hours sooner than six-hourly inspection. Below 800, hourly caught nothing. medium.com/@tinholt/when-wat…
40
Thirteen disciplines of AI engineering by my count, in four layers with one control plane. Seven decide what an agent can do; six decide whether anyone can trust it. Scaled programs stall on the six, because nobody owns them. medium.com/@tinholt/engineer…
1
1
213