Rung 2 · ₹6–12L
Workflow Sprint
Four to six weeks, one named workflow, built with your people. They own the code.
A Workflow Sprint is a four-to-six-week engagement that builds a working agentic solution for one named workflow, jointly with three to five of your own people, who keep the code. It ships an evaluation harness and a measured before-and-after against the baseline, and is priced ₹6–12L.
- Four to six weeks, one named workflow, built with three to five of your own people.
- You own the code. It is not a hosted black box rented back to you.
- Ships an evaluation harness so you can tell regression from variance after we leave.
- It follows the Diagnostic; without a named workflow and baseline, we do not start.
- Priced ₹6–12L, indicative; the next rung is the Capability Cohort.
- Rung
- 2
- Duration
- 4–6 weeks
- Price
- ₹6–12L
- Shape
- 4–6 weeks · one named workflow · built jointly with 3–5 of your people · ships working code plus an evaluation harness
Indicative. Priced per engagement.
Who this is for
A transformation owner or engineering leader who has already run an Adoption Diagnostic, has a named workflow with a baseline, and wants a working agentic solution built and shipped alongside their own people, not a proof of concept that stalls, and not a black box a vendor keeps. This rung builds the thing the Diagnostic pointed at.
What you get
- A working agentic solution for one named workflow, running against your real constraints.
- An evaluation harness that scores the solution on frozen cases before and after every change.
- Three to five of your own people who built it with us and can maintain and extend it.
- A measured result against the baseline set before the build, not a self-reported speed-up.
- A defined human-in-the-loop boundary: what the system proposes and what a person still decides.
Deliverables
The working solution and its code
The full repository, owned by you. Not a hosted black box, not a licence: the code your team built with us and can change.
Evaluation harness
A versioned test suite of real cases that scores the solution on every change, so you can tell regression from variance after we leave.
Runbook and failure-mode notes
How to run and extend it, and a written description of the cases it does not handle, so nobody discovers the boundary in production.
A measured before-and-after
The workflow's baseline metrics and their post-build values, using the definitions agreed at the start.
What this rung will not do
We name what we refuse to sell before we start. The honesty constraint is the product.
- No engagement without a named workflow and a baseline. If the Diagnostic did not produce one, we go back a rung, not forward.
- No black box. You own the code; we do not retain a solution and rent it back to you.
- No unattended automation of steps that carry legal or financial commitment. Those keep a human in the loop by design.
- No cohort or broad training folded in. Training is a later rung and a different sale, not a sprint add-on.
- No logo-for-discount trade. The price reflects the work, not your brand's marketing value to us.
Workflow Sprint, week by week
Text description of this diagram
A four-stage left-to-right sequence for a Workflow Sprint: Week 1 shadow and scope, Week 2 design the agent, Weeks 3 to 4 build together, Weeks 5 to 6 ship and hand over. Arrows connect the stages, showing a four to six week engagement moving from discovery to a shipped agentic workflow the client owns.
Frequently asked questions
It is a four-to-six-week engagement that takes one named workflow (identified and baselined in a prior Adoption Diagnostic), and builds a working agentic solution for it, jointly with three to five of your own people. It ships real code that you own, an evaluation harness that scores the solution on every change, and a measured before-and-after against the baseline. It is a build, not a proof of concept and not a slide deck.
You do. The full repository is yours, built alongside your people so they can maintain and extend it. We do not retain the solution and rent it back as a hosted service. A build that only we can change is a build that fails six months after we leave, so ownership by your team is the point, not a concession.
Because a sprint needs a named workflow and a baseline to build against and to measure, and the Adoption Diagnostic is what produces them. Starting a build without them is how organisations join the 95% of GenAI pilots that showed no measurable return. If you arrive without a named workflow, we go back a rung rather than build on a guess.
Indicatively ₹6–12L, quoted per engagement against the complexity of the workflow. Engagements are billed 50% upfront and milestone-billed thereafter. The price is an opening hypothesis; it reflects the build work, not the marketing value of your logo, which is never traded for a discount.
Any step that carries a legal or financial commitment. The sprint designs an explicit human-in-the-loop boundary: the system proposes, assembles evidence and flags, and a person decides on anything binding. That boundary is written into the runbook so a later change cannot quietly remove it, and it is a permanent design decision, not a temporary phase-one limitation.
Not sure which rung fits?
Start with a free AI Reality Check. Ninety minutes, on-site, one real workflow built live.