An in-house AI capability is not one hire. The credible minimum is an engineer who can build harnesses, someone product-shaped to own workflows, and a fraction of a data person: realistically $400k to $700k a year loaded, plus the two quarters it takes them to learn your business and make the mistakes a specialist has already made on someone else's budget. That investment is correct once AI is core to your product or you have a portfolio of proven workflows to run.
A competent partner lands in the $10k to $50k per month range depending on scope, ships from a pattern library instead of from scratch, and is disposable the day it stops earning its fee. The honest weakness: partners leave, and if the engagement was structured as "they build, we watch," the capability leaves with them.
Phase one, prove (external-heavy, 60 to 120 days). A partner ships two or three measured workflows end to end, installs the harness, the evaluation habit, and the governance page. Deliverables are working systems plus documents your team can run.
Phase two, transfer (mixed, the next quarter). You hire or appoint one internal owner. The partner's job description changes from building to teaching: pairing, documentation, handoff per the deployment rule that the business owns every build.
Phase three, scale (internal-heavy, ongoing). Internal owns the portfolio and the roadmap; external returns for spikes: a new vertical, a hard integration, an audit. The partner headcount goes to zero between spikes, which is the point.
The pattern fails only when phase two is skipped, which is also the most common way it is sold. Insist on the transfer phase in the contract.
Phase one: external partner ships two or three measured workflows and installs the operating patterns. Phase two: one internal hire; the partner switches from building to teaching, with handoff in the contract. Phase three: internal runs the portfolio; external returns for spikes only. Skipping phase two is how capability walks out with the vendor.
Ask to see the harness, not the demo: what can this agent touch, and how is it evaluated? Ask for the golden set from a past engagement. Ask what they refused to automate for a client, and why. Ask how their builds are handed off, and to whom, and what documentation survives them. Ask what the client's monthly token bill was, and how they knew it was worth it. Ask which model they would swap to if their primary vendor doubled prices tomorrow.
A strong partner answers all six with specifics and artifacts. A demo-driven one changes the subject to capabilities. The questions work because they interrogate operations discipline, which is where AI engagements actually succeed or fail.
Not the technology; five numbers. Which workflows run on AI. What each costs per unit and returns per unit. Who owns each one. What the evaluation caught last month. What the portfolio spends and saves in total. A partner or team that cannot produce those five numbers on one page, monthly, is running pilots on your payroll, whatever the deck says.
Sequence it: a partner proves the first workflows and installs the patterns in 60 to 120 days, one internal owner takes over as they stabilize, and specialists return for spikes. Hire a full team only once AI is core to your product or portfolio.
Competent partners typically run $10k to $50k per month and ship from proven patterns. A credible in-house minimum (builder, workflow owner, fractional data) is $400k to $700k a year loaded, plus two quarters of ramp.
Interrogate operations, not demos: their harness design, a real golden set, a refusal story, their handoff process, a client's token economics, and their model-swap plan. Specifics and artifacts distinguish operators from presenters.
Bring one AI question from this article. We will tell you what we would do, whether or not you hire us.
Book a 15-Min Chat →