We do not sell hours, so we cannot price them. Every BearPlex engagement is scoped to an outcome: a diagnostic with a fixed scope, a build priced per pod, or a retainer that keeps a live system honest. The number comes after we understand the problem, and it arrives fixed.
Hourly billing structurally rewards inefficiency: the longer the work takes, the more the vendor earns. We refuse to operate that way.
Fixed after diagnostic
Every engagement starts as a diagnostic. The number for the build arrives fixed, with the architecture that justifies it.
You own everything
Code, infrastructure, and IP are yours from the first commit. The price never includes a hostage clause.
Which shape fits
Three shapes. One commitment.
Every engagement starts as a diagnostic. The build and guardian phases follow only if the architecture warrants them. Find your situation; the shape follows.
You know the problem, but nobody has sized the build.DiagnosticFixed fee
The architecture is under audit pressure and needs a map before more code lands.DiagnosticFixed fee
A regulated production system has to ship from zero.BuildPod-based
The blueprint exists and needs an engineering pod to execute it.BuildPod-based
The system is live and you cannot afford regression after launch.GuardianRetainer
Someone has to own the SLOs and the quarterly architecture review.GuardianRetainer
01Fixed fee
Diagnostic
A two-week discovery sprint. We deconstruct the problem to first principles and produce the architecture before a single engineering hour ships.
Ends with scope, phasing, and a fixed price for the build. If the build is not worth doing, the diagnostic says that too.
02Pod-based
Build
A 12 to 16 week engineering pod executes the core platform. Modular, evidence-driven, examiner-ready by the end of the engagement.
Ships a production system in your VPC or sovereign cloud, with the evidence packet and the operations handover.
03Retainer
Guardian
Ongoing partnership. We own the SLOs, run the RLHF loops, and ship the quarterly architecture reviews that keep the system compounding.
Holds the line after launch, on a monthly retainer, for operators who cannot afford regression.
By family
Where the number comes from.
The pricing language differs by family, but the route to it never does: the page describes the work, the diagnostic sizes it, the quote arrives fixed.
AI engineering
Scoped through the diagnostic, then quoted as fixed milestones. Each page describes what an engagement covers.
Honest ranges, published so you can budget before we ever talk. Every floor is where credible scope starts. Your exact number still arrives fixed, after the diagnostic.
The honest part first: for most teams, buying the SaaS is the right call. Custom software earns its price in a narrow set of cases: when the off-the-shelf tool bends your core workflow, when per-seat pricing scales faster than your revenue, or when the vendor holds data you cannot afford to leave behind. If a subscription solves your problem, we will say so on the first call.
Project builds
Fixed price after the diagnostic. The floor is where credible scope starts; most projects land inside the typical band.
Marketing website (WordPress)
from$2,000
typical $3,000-$8,000
Page count and custom templates
Content migration volume
Forms, CRM, and analytics integrations
Performance and SEO hardening
The right call when the site is content-led and your team will edit it themselves.
Marketing website (Next.js)
from$4,000
typical $6,000-$15,000
Custom design system versus adapted template
Headless CMS integration
Animation and interaction depth
Performance and SEO targets
For brands where load speed and design precision are the point.
E-commerce storefront
from$4,500
typical $8,000-$25,000
Catalog size and variant complexity
Payment, shipping, and tax integrations
Checkout customization
Inventory and ERP sync
Platform-first when Shopify or WooCommerce fits; custom when your margin logic does not.
SaaS MVP / custom platform
from$15,000
typical $25,000-$80,000
User roles and permission depth
Third-party integrations
Compliance and audit requirements
Real-time and multi-tenant features
Priced to reach paying users, not to gold-plate a product nobody has validated.
AI system / agent build
from$15,000
typical $25,000-$75,000
Data readiness and access
Tool surface the agent operates
Evaluation and golden-set work
Guardrails and human review loops
The demo is cheap. The production system with evals and guardrails is the price.
Custom internal tool / CRM / ops system
from$15,000
typical $25,000-$70,000
Workflow and approval complexity
Migration from spreadsheets or legacy tools
Integrations: accounting, email, telephony
Reporting depth
Only worth building after the off-the-shelf tool has actually run out of road.
Teams and ongoing care
BearPlex engineers work AI-augmented, so a monthly rate is judged by what ships each month, never by hours logged.
Embedded engineer (AI-augmented)
from$3,000/monthper role
Seniority and specialization
Timezone overlap you need
Scope of ownership
One AI-augmented engineer, embedded in your team and measured on shipped output.
Embedded pod (2 senior devs + PM + QA)
$9,500/month
larger pods range to about $20,000/month
Pod composition and size
Product maturity and codebase state
Delivery cadence you expect
A whole delivery unit shipping as one: two senior engineers, a PM, and QA.
Care plan / maintenance
from$150/month
typical $350-$2,000/month
Update and patch cadence
Response time you need
Monitoring and uptime scope
Volume of small changes
Keeps a live system patched, monitored, and honest after launch.
The estimator gives you a working range in minutes. The call gives you a real number.
FAQ
Why there is no rate card.
The honest version, asked the way buyers ask it.
Hourly billing structurally rewards inefficiency: the longer a project takes, the more the vendor earns. We refuse to operate that way, so there is no hourly rate to publish. Our compensation is tied to outcomes the client cares about, not hours we burn.
Every engagement starts as a diagnostic: a two-week discovery sprint that produces the architecture, the phasing, and a fixed price for the build. Paid discovery gives us the time to understand your actual problem and quote a real price instead of an optimistic estimate.
If we underestimate scope, we absorb the cost overrun. That is the point of fixed-price milestones: the risk of inefficient execution sits with us, not with you. Genuine scope expansion, where you add new requirements mid-engagement, is re-scoped and re-priced as a new milestone.
Start with the diagnostic
Bring the problem. We bring the number.
A two-week diagnostic ends with the architecture, the phasing, and a fixed price for the build. If we should not take the work, we will tell you that too.