Skip to content
PRANINNOVATIONSProduction Specialists

Pricing

Published, because the people we want to work with ask early

Bands and the reasoning behind them. Not a rate card, and not a number that appears only after three calls.

01

Vision Feasibility Check

Free

written memo, no call

Send 30–50 sample images. Get a written answer on whether this works.

Send sample images
  • Realistic accuracy band for your classes
  • Hardware class the frame rate would need
  • Risks visible in your images
  • Go, no-go, or fix-this-first
  • Returned in a few working days
  • No call required at any point
02

AI Pilot

$28K–60K

fixed price, four weeks

Signed acceptance criteria, working software, and a costed path to production.

Scope a pilot
  • Acceptance criteria signed before any code
  • Working software against a real slice
  • Evaluation report against those criteria
  • Reference architecture
  • Costed, phased path to production
  • Or a written case for not proceeding
03

Production Build

$75K–180K

indicative, 8–14 weeks

The system itself, phased, quoted per phase before that phase starts.

Get a costed plan
  • Numeric acceptance criteria throughout
  • Deployed in your environment
  • Evaluation report per phase
  • Weekly written progress
  • Runbooks and administrator training
  • Source, weights and IP assigned to you

The full ladder

Everything, with what moves it

Where an engagement lands inside a band is set by the six factors below, not by how the conversation goes.

Vision Feasibility Check
Days · sample images in, memo out
Free written memo
Feasibility Study
1–2 weeks
fixed, from $9,500
AI Pilot
4 weeks · signed acceptance criteria
fixed, $28K–60K
Production Build
8–14 weeks
indicative, $75K–180K
Vision System, single line or cell
8–14 weeks · includes hardware and commissioning
indicative, $60K–150K
AIoT Fleet Deployment
10–20 weeks · scales with fleet size
indicative, $90K–300K+
Enterprise AI System
3–6+ months
scoped, $180K–500K+
Embedded AI Team
3+ months · named engineers, itemised delivery
from $28,000 / month
Managed Edge & Model Fleet
Ongoing · drift, retraining, OTA, device health
from $4,500 / month

Six price movers

What actually moves the number

01

Data readiness

Whether labelled examples exist, whether the labels agree with each other, and how much capture has to happen before training can start.

02

Integration depth

Writing to a controller on a live line costs more than posting to an API. Every system the result has to reach adds interface work and a change window.

03

Reliability bar

A system that advises a supervisor and a system that fires a reject gate carry different fail-safe design, different testing, and different sign-off.

04

Evaluation burden

Regulated and safety-critical work needs a documented validation record, held-out sets assembled with your team, and evidence a review board will accept.

05

Deployment constraints

Air-gapped sites, controlled technology, hostile environments and multi-site rollouts all add commissioning time that a cloud deployment never sees.

06

Usage economics

Per-tenant metering, inference budgets and the pricing model of the product being built shape the architecture, and therefore the build.

Start small enough to measure. Scale only when the evidence supports it.

FAQ

On the numbers

Because the alternative wastes everyone's time. You should know inside a minute whether we are in your range, and we would rather lose an unqualified conversation early than a qualified one late.

The Feasibility Check is free. The Feasibility Study and the AI Pilot are fixed once scope is signed. Everything above them is indicative until discovery has happened, then quoted per phase before that phase starts.

Pricing far under the US market is a signal, and not a good one: it means a corner is being cut somewhere, usually in testing, integration or who actually does the work. We sit modestly under boutique rates and say what the money buys.

Third-party costs are billed to your accounts at cost, never marked up through us: cloud, cameras, edge hardware, lighting, enclosures, mounting and any licences the deployment needs.

It includes a site survey, hardware, physical installation, integration with equipment you already run, and a monitored production period before the system is allowed to act. That is engineering time on your floor, not just in a repository.

Start with a call, then a costed plan

Thirty minutes on the problem, the site and the constraints. If it looks like a fit, the next step is a four-week AI Pilot at a fixed price, with acceptance criteria signed before any code is written.