Pricing

Four ways to engage

Modern Team Augmentation

AI-enabled, dedicated engineers embedded in your team for the long term.

Platform
Dap included
Billing
Human hours + unattended agent tokens
Suitable for:
Clients who want top engineering talent and want to get the most out of what agents can do.
Select plan

Classic Staff Augmentation

Dedicated engineers embedded in your team for the long term.

Platform
Dap not included
Billing
Human hours
Suitable for:
Clients who aren’t ready for AI in delivery, or can’t use it. Same rules as the old days.
Select plan

Elastic Projects

AI-enabled team, sized and flexed by Crunchloop to fit an open-ended, evolving scope.

Platform
Dap included
Billing
Human hours + unattended agent tokens
Suitable for:
Clients who trust Crunchloop to own delivery end-to-end, and who get the most value from work agents can run unattended.
Select plan

Fixed Projects

AI-enabled team, sized by Crunchloop to deliver one defined, bounded effort.

Platform
Dap included
Billing
Fixed price by scope estimation
Suitable for:
Clients who need a well-defined deliverable with a hard budget — pilots, POCs, architectural reviews, refactors.
Select plan
See what these look like for real clients.Check our Work
See how they play out in practice.See Use Cases

Common paths for real-world clients

  1. Diagnose

    — see what’s already there, or that there’s nothing built yet.

    • Fixed Project
  2. Set up

    — close the gaps so an AI-assisted team can actually run.

    • Fixed Project
  3. Ready to scale

    — decide how to continue building, depending on what fits for you.

    • Modern Team Augmentation
    • Elastic Projects

Before the loop can run, something has to crunch.

The principles behind our plans

1. Variable cost protects quality on open-ended work.

If price and deadline are fixed before the scope is fully known, something has to give when reality doesn’t cooperate. Experience tells us that it’s your quality, or your scope, that quietly gets cut to make the deadlines. Not bad faith — just the math of the constraint.

Scope, time and cost shape project quality.

“Responding to change over following a plan.”

The industry already learned this, decades before AI. It’s why Agile replaced Waterfall. AI doesn’t change that math. It makes implementation faster — requirements keep changing regardless.

Fixed price still makes sense for well-scoped work. For open-ended engagements, locking in a number ahead of time is exactly the mistake this industry already spent decades learning to avoid.

2. Measuring both kinds of effort maps reality.

It’s not just human hours anymore. Agent work is real, and it counts too — which means there are two real costs in AI-assisted delivery now, not one. Measuring both, honestly, is disruptive: it’s not how software work has traditionally been priced.

Measurement

We’re doing it anyway, because we believe it’s the only way this holds up at scale.

We don’t know exactly how the balance between attended and unattended AI work will settle as this keeps evolving — nobody does yet. The real challenge here isn’t picking a number. It’s building a framework that actually fits what these new realities need — and we’re building it from conviction, not convenience.

3. The platform is what protects quality while everything else accelerates.

AI lets teams ship more, faster. But nothing about that automatically keeps quality checks moving at the same pace — so quality can quietly fall behind, even while output looks great on paper. That’s not inevitable, though — it’s what happens without the right platform underneath. With one, speed and quality move together instead of trading off. That’s what Dap is for.

Comparing two people who share the same traits, environment, and processes, the one with higher AI adoption reports higher levels of software delivery instability. AI adoption not only fails to fix that instability — it’s currently associated with increasing it.

DORA, State of AI-assisted Software Development(2025), p. 38 (Figure 28, 89% credible intervals) and p. 41

AI adoption has a negligible effect on organizational performance when platform quality is low, but a strong, positive effect when platform quality is high — an investment in AI without a corresponding investment in high-quality platforms is unlikely to yield significant returns at all.

Why Dap isn’tself-serve

Dap isn’t something you sign up for on your own. It started as the internal layer our own engineers built to run our own client work, long before it had a name — and having our team manage it is what makes it work at its best.

Talk to us about what fits.