Services / sprint / automation
Prompt & Eval Library Setup
Stop versioning prompts in Notion. Treat them like code.A one-time install: your prompts versioned in git, evals running in CI, and an A/B harness so prompt changes ship like any other deploy. Two weeks. Hand-off to your team after.
price
from $3,500
timeline
2 weeks
cadence
one-time
scope
One-time / fixed scope
Where this fits in the services matrix.
Every service page now names the buyer state, the commercial shape, and the next route. That keeps the catalog navigable instead of feeling like disconnected offers.
Build automation with a fixed scope and written handoff.
from $3,500 · 2 weeks · One-time / fixed scope
Use the diagnostic or book a call to confirm fit before scope is written.
Not sure this is the right service? Run the route finder and get the matching path.
The offer is a route, not a loose task list.
This diagram gives every service page a concrete operating model: intake, system design, implementation, proof, and handoff.
service operating path
Surface ⇄ System
Prompt Library flow
The diagram is intentionally simplified: it shows the buying logic and operating path, not a decorative fantasy architecture.
price
from $3,500
timeline
2 weeks
cadence
one-time
The outcome, not just the output.
- 01Prompts in git with diff history, not in a Notion doc
- 02CI evals on every prompt PR
- 03A/B harness for safe prompt rollouts
- 04Documentation pattern your team will actually use
Concrete artifacts you keep — and what we leave out.
Working code, written docs, dashboards your team owns. We also list what this engagement deliberately does not cover, so scope is honest before you click.
- Prompt repo structure + naming conventions
- Eval suite — at least 3 quality dimensions per prompt
- GitHub Actions workflow that runs evals on PR
- A/B rollout helper (e.g., Statsig / LaunchDarkly / hand-rolled)
- Migration of up to 20 existing prompts
- Team training session (60 min Loom + live Q&A)
- Authoring brand-new prompts (we move what you have)
- Long-running eval maintenance (see AI Quality Retainer)
How the engagement actually runs.
- 1Day 1–4
Repo + conventions
Set up the prompt repo, naming conventions, frontmatter schema, version pinning approach.
Prompt repoCONTRIBUTING.mdNaming spec - 2Day 5–8
Evals + CI
Author baseline evals, wire into GitHub Actions, fail-on-regression thresholds.
Eval suiteCI workflowThreshold config - 3Day 9–12
A/B + migration
Stand up the A/B helper, migrate up to 20 existing prompts, run them through evals to get a baseline.
A/B helperMigrated promptsBaseline eval report - 4Day 13–14
Training + handoff
60-minute training session, recorded Loom, written runbook for adding prompts/evals/tests.
Training LoomRunbook
Receipts, not promises.
- 2 weeks
- Setup to handoff
- 20
- Prompts migrated
- baseline
- 3+
- Eval dimensions per prompt
Common questions.
01We use LangChain prompts — does that work?
02What if a prompt change breaks evals?
Ready to start Prompt Library?
A 30-minute call to confirm fit, scope, and timeline. No pressure, no slides.
automation system
From offer to operating system.
Prompt & Eval Library Setup is presented as a real engagement, not a generic service page: the surface, backend shape, delivery artifacts, and conversion path are all visible before the first call.
Scope Prompt Libraryprice
from $3,500
timeline
2 weeks
tier
B
Living architecture
Scope ⇄ Ship
The page now exposes how the engagement moves from buyer pain to production artifact, then into measurement and next-step routing.
Scope Prompt Library- 01Repo + conventionsSet up the prompt repo, naming conventions, frontmatter schema, version pinning approach.
- 02Evals + CIAuthor baseline evals, wire into GitHub Actions, fail-on-regression thresholds.
- 03A/B + migrationStand up the A/B helper, migrate up to 20 existing prompts, run them through evals to get a baseline.
- 04Training + handoff60-minute training session, recorded Loom, written runbook for adding prompts/evals/tests.
Conversion path
Surface ⇄ System
01
Diagnose
Confirm the real automation constraint, current surface, and business goal before writing code.
02
Design the system
Turn the offer into screens, data, workflows, ownership boundaries, and a measurable delivery plan.
03
Ship the artifact
Deliver Prompt Library as working code, docs, dashboards, or launch assets your team can actually use.
04
Route the next move
Decide whether the work becomes a one-time delivery, a care plan, or a larger product build.
Proof assets
Real only
Asset slot
Service proof visual
Add a real screenshot, deliverable preview, or dashboard capture from a shipped engagement when approved.
Verified asset
Founder/operator photo
Real founder photo reinforcing principal-led delivery.
Asset slot
Client quote or logo
Add only permissioned testimonials or logos tied to this service category.