Start a project | Dualis Logic

Who you are

Name

Work email

Organisation

Your role

Sector and frameworks

Sector

Every framework that binds your infrastructure. A BAA or a cardholder data environment changes where the model may run.

What is in the way

Token costsThe harness resends large tool schemas and long context every turn.Tool-calling failuresWrong arguments, malformed JSON, calls that break your API contracts.Latency and rate limitsMulti-turn loops stall when the API throttles or slows down.Data that cannot leaveYour rules keep source code, schemas or records away from third-party endpoints.Engineering backlogYou want an agent harness with human review gates.More than one, or not sure yet

Volume and spend

Monthly AI API spend, current or projected. The bands sort the work; none of them promises a saving.

Under $2,000 a monthCaching and prompt work come first.$2,000–$7,500 a monthThe audit decides. Check caching first.$7,500–$25,000 a monthWorth pricing a model against caching.Over $25,000 a monthA strong candidate, if your tools are stable.Not measured yetThe audit measures it.

Tools per agent

Optional. A guess is fine.

Turns per task

Optional. The average is enough.

The engagement you have in mind

We publish the prices, so you can already answer this.

Token & Throughput Audit$4,500, one weekCompliant AI Readiness Assessmentfrom $12,000, 2–3 weeksThe Governed AI Workspacefrom $28,000, 3–5 weeksThe Factoryfrom $38,000, 4–6 weeksCustom Tool-Calling Modelsfrom $45,000, 6–10 weeksFractional AI Engineering Leadfrom $9,000/mo, three-month minimumMore than one, or not scoped yet

Timeline

This quarterNext quarterDriven by a fixed dateAn audit, a board milestone, a launch.

Your architecture

Your stack, and what success looks like

Specifics are better than polish. If there is a deadline, name it.

We reply within two business days, including when the answer is no.

Five minutes of technical detail beats an hour of introductory calls.