AI development pricing and rates
Our published rates, the arithmetic behind them, and an honest comparison with the alternatives. For market-wide numbers rather than ours, read what AI development costs in 2026. To choose between the two models, compare staff augmentation and project outsourcing.
A dedicated AI engineer costs $2,800 per month, full-time, rolling monthly. Fixed-scope project builds start at $4,500 and most land between $6,000 and $26,000 depending on integration count and evaluation rigour. Model API spend and cloud hosting are billed to your accounts at cost, never resold with a markup. No annual contract on anything.
Three ways to buy
Dedicated engineering pod
MONTHLYBest for: Teams with a roadmap and no in-house AI experience. Capacity you can turn up and down monthly.
- One full-time AI engineer, 40 hours a week, not shared with other clients
- Onboarded within 48 hours
- Daily written reporting, weekly call if wanted
- Senior architecture review on everything the pod ships
- Rolling monthly, 30 days notice, no annual contract
Project build
FIXEDBest for: A defined outcome with a deadline. You want a system delivered, not a team to manage.
- Free audit and strategy session before you commit
- Fixed price and fixed date from a written scope
- Evaluation harness and regression suite shipped with the build
- Launch support, documentation and runbooks
What typical projects cost
Ranges, not point estimates, because the same brief can differ by a factor of three depending on how many systems it has to touch. The low end assumes clean data and one integration; the high end assumes messy data, several integrations and a compliance review. Agent builds are broken down further in what AI agents actually cost to build.
| Project | Cost | Timeline |
|---|---|---|
| Single-purpose chatbot on your own content | $4,500-6,000 | 2-3 weeks |
| AI chatbot + knowledge base with handoff | $4,500-5,500 | 3-4 weeks |
| Workflow automation suite | $4,500-7,500 | 4-6 weeks |
| Document search / RAG over one corpus | $5,500-9,000 | 5-7 weeks |
| Multi-tenant RAG with isolation and SSO | $12,000-20,000 | 8-12 weeks |
| Custom agent platform | $14,000-26,000 | 10-16 weeks |
| AI readiness audit, written | $2,000-3,500 | 1-2 weeks |
| Evaluation harness for an existing system | $3,000-6,000 | 2-3 weeks |
What drives the number up or down
In rough order of how much they move the total. None of these is the model you pick - that is almost never the expensive decision.
Integration count
Each external system is an auth flow, a rate limit, a data model to learn, a sandbox to get access to and a failure mode to handle. One integration is a week. Five is not five weeks - it is more, because they interact. This is the single biggest driver and it is the one most estimates miss.
Data quality
Clean, consistently structured documents in one format are cheap. Scanned PDFs from 2014, six naming conventions and three systems of record that disagree are expensive - often half the project. Send us ten real, representative files at scoping and the estimate gets dramatically more accurate.
Evaluation rigour
A demo needs no eval suite. A system people rely on needs a labelled test set, regression runs on every prompt change, and agreed thresholds for shipping. That is typically 15-25% of a build, and it is the part that decides whether the thing still works in month six.
Compliance burden
GDPR scoping, data residency, audit logging, per-tenant isolation and a DPA add real engineering - not paperwork. HIPAA more so. Budget 20-40% on top of the equivalent unregulated build.
Latency targets
Under 3 seconds is usually free. Under 1 second means caching, re-ranking choices, smaller models on the hot path and a lot of measurement. If sub-second is a hard requirement, say so at scoping - it changes the architecture, not the polish.
Volume
Cost scales with tokens and documents, not with users. Ten thousand documents and ten million behave differently at the embedding, storage and re-ranking layers. Tell us the real number, including the growth you expect.
How we compare - including where we lose
| Option | Typical rate | Best for | Trade-off |
|---|---|---|---|
| US agency | $150-300/hr | On-site enterprise work, US-only data handling, procurement that requires a domestic vendor | 3-5x the cost for the same engineering |
| EU agency | EUR 120-250/hr | EU data residency and an EU legal entity on the contract | Similar cost premium, narrower AI bench |
| Freelancer | $60-150/hr | Small scoped tasks under $5,000, or a specific skill for two weeks | No bench, no continuity, no cover when they take a holiday |
| In-house hire | $180-260k/yr loaded | Long-term ownership of an AI roadmap and mentoring the next hires | 3-6 months to fill, 4-8 weeks to ramp |
| MacroCoderz | $2,800/engineer/month, projects from $4,500 | Production AI systems and ongoing capacity without a hire | No on-site presence. |
If your work is genuinely on-site, or your procurement rules require a domestic vendor, a US agency is the right answer and we will not pretend otherwise. Under $5,000 of scoped work, hire a freelancer. We are the right choice when you need production AI engineering, at volume, from people who have shipped it before. Agencies reselling AI work to their own clients should start at white-label AI development.
What is not included
We keep these out of our invoices on purpose. Reselling API credit with a margin creates an incentive to use more tokens than the problem needs, and it leaves you without the vendor relationship when the engagement ends.
- Model API spend - OpenAI, Anthropic, Cohere and similar, billed to your account at cost. A typical production RAG system runs $150-800 a month.
- Cloud infrastructure - hosting, vector database, storage. $50-400 a month for most builds, more at scale.
- Third-party licences and SaaS seats your system depends on.
- Design as a separate discipline. Engineers build usable interfaces; they are not a brand or product design team.
- Ongoing maintenance after handover, which is quoted separately or covered by a pod.
Frequently asked questions
One month for a pod, $4,500 for a project build. Below $4,500 the scoping and handover overhead outweighs the build and a freelancer will serve you better - we will tell you that rather than take the work.
If it overruns because we underestimated, we absorb it - that is what fixed price means. If the scope changes, we quote the change in writing before doing the work, so nothing appears on an invoice you did not approve.
No, and deliberately so. API spend, cloud hosting and third-party licences are billed to your own accounts at cost. We do not resell credit with a markup, and you keep the vendor relationship when the engagement ends.
Pods are rolling monthly with 30 days notice. Project builds run until handover. There is no annual lock-in on either.
No. The strategy session is free.
Get a fixed number for your project
A free strategy session, a written scope and a fixed quote.