Engagement · Build & deploy

Ship your first AI system. End to end.

We take one opportunity — usually the top pick from an Audit — and build, deploy, and hand over a production AI system. Plugged into your real stack. With observability, guardrails, and a playbook your team can run without us.

8–12 weeksEngagement
From $15,000KES 1,935,000
1 systemIn production
Start here if…
  • You already know what you want to build (or completed an Audit).
  • You have a real process with real data — not a hypothesis.
  • You need a working system, not a proof of concept that dies in a demo folder.
  • You want your team to own the system at handover — not be locked in to us.
  • You care as much about observability, cost, and compliance as about the "wow" of AI.
A production system. And everything around it.
We ship software. But software alone isn't the deliverable — it's one of seven.
01 / System

Production AI system

Live, in your infrastructure (cloud or on-prem), integrated with your real data sources. With authentication, rate limiting, and cost controls from day one.

02 / Guardrails

Safety & accuracy guardrails

Input validation, output filtering, hallucination checks, escalation paths, and content policies — tuned to your compliance posture.

03 / Observability

Full observability stack

Metrics, structured logs, traces, cost dashboards, quality evals, and alerts. You can see exactly what the system is doing, how well, and for how much.

04 / Eval suite

Continuous evaluation harness

Automated test cases covering accuracy, bias, latency, and cost regressions. Your team can change prompts or models without flying blind.

05 / Runbooks

Operational runbooks

Incident response, model updates, data drift handling, cost optimisation. A binder of procedures, not a hope.

06 / Training

Team training & handover

Live sessions with engineering, operations, and stakeholders. Code walkthroughs. Three post-launch office hours. You leave the engagement capable of extending the system yourself.

Six phases. Ten weeks typical.
1
Scope

Success criteria, non-goals, data contracts.

2
Design

Architecture, model selection, safety spec.

3
Build

Iterative development in your staging env.

4
Evaluate

Accuracy, latency, cost, bias benchmarking.

5
Ship

Production deploy. Monitoring live.

6
Handover

Training, runbooks, post-launch support.

What production AI looks like.
60–80%

Reduction in manual-work hours for the targeted workflow

< 2s

P95 latency for user-facing interactions

99.5%

Uptime target, monitored end-to-end

4–9 mo

Typical payback window

Integration questions.
No, but it helps. If you haven't, our first two weeks include a compressed audit-style scoping phase to make sure we're building the right thing.
You do, unconditionally. Code in your repository, models in your cloud accounts, data in your databases. No vendor lock-in to us.
Whatever fits. We're tool-agnostic — OpenAI, Anthropic, open-weight models (Llama, Mistral, Qwen) on cloud or self-hosted, Google, Cohere, local small models. Choice is part of the design phase, not assumed.
We design for whatever your regulator requires. Includes on-prem deployments, self-hosted open-weight models, regional cloud, and documented data-flow diagrams.
We plan in 2-week iterations, so scope can adapt. Major pivots trigger a formal change order — no surprise invoices.
Three post-launch office-hours sessions are included. Beyond that, retainers (4h, 8h, 16h/month) are available if you want ongoing support. Most clients don't need one after month two.
30% on kick-off, 40% at midpoint milestone, 30% on handover. Paystack, bank transfer, USD wire. 3-month and 6-month installment plans available for larger engagements.
The smallest engagements are scoped to one well-defined workflow with ready data. Complex multi-system work or data-pipeline builds start higher. Full transparency on the pricing page.
Integration engagements start at $15,000 Scope-driven. We'll give you a fixed quote after a 30-minute scoping call — not a vague range.
See full pricing Start scoping
Get a fixed quote in 30 minutes.
Tell us what you want to build. We'll scope it on a call, then come back with a written plan, fixed fee, and timeline.
Response within 24 hours
Fixed-fee engagements
Your code, your models