Like the large engineering firms, we don't publish a rate card: every project is priced after we understand the process. What we do publish is how to engage us, what each model includes and how we get to a number.
Possition works under four engagement models priced in USD: a fixed-fee discovery sprint, fixed-price delivery by milestone, a dedicated nearshore pod billed monthly, and a managed service for AI systems in production. On select engagements, part of the fee is tied to a verified KPI. Contracts are signed with NEXIA LLC (Delaware, USA).
Almost every engagement starts with a discovery sprint. After that, the model depends on how well defined the scope is and whether we run the system for you.
To start: before you commit a large budget.
Two to three weeks to map the process, audit the data and systems, design the architecture and build the cost model with payback, NPV and alternatives. It ends with a fixed-price proposal, or with a recommendation not to automate.
When the scope is clear: one agent, one automation, one integration.
Price is fixed per milestone and Possition carries the estimation risk. Each milestone ships working and is accepted before it's invoiced. Scope changes are priced and approved before any work starts.
When the roadmap evolves or there are several systems to build.
A cross-functional team (tech lead, engineers, data and ML, QA) working your hours and your priorities. It's capacity, not loose hours: scale up or down with notice, and every sprint ends with a demo.
When the system is in production and you want someone accountable for it.
We run your agents, models and automations in production: monitoring, automated evaluations, prompt and model tuning, knowledge-base updates, cloud and AI cost control and a monthly KPI report, under a service-level agreement.
On select engagements, part of the fee is tied to a verified KPI: hours freed, cost per resolution, accuracy or conversion. The baseline is measured in the discovery sprint and the result is verified against production data. It's the outcome-based logic large consultancies use, applied to mid-sized projects.
These factors explain almost all the difference between one proposal and another. We measure them in the discovery sprint, before we set a price.
A system with a documented API is predictable work; a legacy one with no API or documentation can take longer than the new system. It's the factor that moves a proposal the most.
Clean data is used as is. Data spread across spreadsheets with different formats and duplicates needs a normalization phase first.
A system that informs costs less than one that takes action. The more it decides on its own, the more validation, permissions and oversight it needs.
Role-based access, an audit trail for every operation and data-protection requirements (for example SOC 2 or HIPAA controls) add real work.
Covering the main flow is one thing; covering every exception is another. Defining where scope ends is part of the proposal.
Every fixed-price delivery comes with a warranty: if anything agreed doesn't work as specified, we fix it at no cost. After the warranty, your team can own the system (documentation and training included) or move it to the managed service.
30 minutes, free, scheduled in US Eastern Time. We check whether there's a business case.
Process map, data, architecture and cost model. It ends with a proposal or a "not worth it".
Engagement model, scope, milestones, team and price in USD. Signed with NEXIA LLC.
Biweekly demos, production launch with monitoring and, if chosen, a managed service.
For the same reason large engineering firms don't: price depends on integrations, data and real scope, and a generic rate card doesn't hold up. We publish how to engage us and give you a concrete number after the discovery sprint.
US dollars. Contracts are signed with NEXIA LLC, our Delaware entity.
A discovery sprint followed by fixed-price delivery. If the system then needs continuous evolution, it moves to a dedicated pod or a managed service.
On select engagements, yes: part of the fee is linked to a KPI measured against the discovery-sprint baseline.
You do. Code, data and trained models stay on your infrastructure, in your name.
Tell us what you need to solve
In a 30-minute discovery call we review the process, the systems and the numbers, and tell you what's worth building, what isn't, and what return to expect. No cost, no commitment.
