Pricing
We publish ranges by engagement, not per-hour rates. Every project ends in a fixed scope with written assumptions before build, so the number you approve is the number you pay.
Estimate your scope
Answers update the range live. Not a quote — an anchor.
What are you building?
How many system integrations?
Reliability bar?
Estimated range
$35k – $74k
USD · fixed on approval · milestone-billed
Best match
AI pilot
The range narrows to a fixed number after a paid architecture sprint. It never widens.
Engagement bands
Ranges reflect real projects we have shipped and the ones we quote today. Every real quote is a fixed number with written assumptions, produced from an architecture, not from a call.
Scoping call
30 minutes
Free
You leave with a written scope, an honest estimate, and our view on whether AI is even the right tool. No sales layer, no obligation.
Architecture sprint
1–3 weeks
$5k – $15k
The design document that makes the rest of the number honest: data flows, model choices, failure modes, and the success measure you will judge us against.
AI pilot
2–8 weeks
$20k – $60k
One narrowly defined workflow, shipped end-to-end with real integrations, an evaluation harness, and a human-in-the-loop gate wherever mistakes are costly.
Full product build
6 weeks – 6 months+
$50k – $250k+
Multi-tenant platforms, full application frontends, or systems where correctness is non-negotiable. Priced from the architecture sprint, not from optimism.
Retainer
Rolling
Monthly, scoped
For products we ship and clients ask us to keep running: monitoring, evaluation, and incremental features. Cancelable, because it must remain earned.
USD by default · payable in PKR or QAR to our registered entities · milestone-based on fixed scope, monthly on retainers
What drives the price
The number of systems the software must touch: CRMs, telephony, databases, third-party APIs. Each seam adds authentication, error handling, and test surface. Usually the largest driver.
An internal triage tool can be wrong occasionally; a customer-facing agent cannot. Raising reliability means evaluation suites, guardrails, and human gates: real engineering, priced accordingly.
Clean, well-labeled data compresses timelines. Messy sources (inconsistent PDFs, undocumented schemas) add work at the ingestion and evaluation stages.
AI features need AI-native UX: streaming states, optimistic actions, graceful failure. Custom design systems and complex flows scale the frontend work.
What doesn’t
At pilot scale, token costs are a rounding error next to engineering time. Anyone quoting a price dominated by “AI costs” is telling you about their margins, not your project.
In a well-architected system, provider is a config decision, not a budget line. We route by task: cheaper models where quality allows, stronger ones where it matters.
One workflow done properly beats three done as demos. We charge for scope that gets to production, not for the width of a wishlist.
FAQ
Because rate cards optimize for the wrong thing: they anchor negotiation on hours rather than outcomes, and they reward moving fast and shallow over doing it right. We quote fixed scopes with written assumptions: the number you approve is the number you pay, and the boundary is exactly documented in advance.
In two honest steps. Discovery ends in a written scope and a ranged estimate; the architecture sprint turns that scope into a fixed quote with written assumptions. Estimating from a design is engineering; estimating from a conversation is guessing.
Changes are priced when they arise, in writing, or they do not happen. No end-of-project surprises, no silent burn on time-and-materials. If a change is a genuinely good idea we will say so, and quote it. If it isn’t, we will say that too.
Occasionally, for a small number of startups where the fit is unmistakable. It is not a discount; it is a co-investment, and we treat it accordingly. Ask on the scoping call; the honest answer is often no.
Yes. That is the default path. Most engagements start with a fixed-scope pilot: one workflow, one success measure, a clean decision point at the end. From there, work continues as a retained build with weekly demos or a hand-off, whichever serves you.
Third-party subscriptions (LLM APIs, telephony, hosting), your own team’s time in demos and reviews, and any data-collection or annotation you commission separately. We list all of these in the assumptions section of the scope, so there are no line-item surprises.
Next step
A scoping call ends in a written scope and a ranged estimate. No rate card required, no obligation.
Prefer async? hello@visionnexera.com · We reply within one business day.