paul.waweru
Discuss a project
Paul Waweru / AI engineering consultant

AI systems.
Built for work.

I build agents, knowledge systems, and automations for startups and growing businesses.

Discuss a project Explore my work
Illustrated portrait of Paul Waweru in a white sweater
Strategy meets
hands-on engineering.pw.

A builder’s perspective.
From the first decision to delivery.

Since 2019Software engineering
Web + mobileEnd-to-end product delivery
FounderArtebello, AI-first consultancy

Good ideas need
working systems.

A selection of the products, architectures, and decisions behind my work.

Construction · SaaS

The Contractors Toolbox

Construction software bringing RAMS, COSHH, and vehicle, plant, and equipment inspections into structured operational workflows.

Visit The Contractors Toolbox
Knowledge systems · MCP

Narravo

An AI brand-intelligence platform connecting reviewed business knowledge to strategy, content workflows, and compatible AI clients.

Open Narravo
Operations · Web & mobile

Civicore

A political campaign operations platform connecting role-based workflows, web services, administration, and a Flutter mobile application.

Visit CivicoreOpen Civicore app
Strategy · Digital transformation

Artebello

A live AI readiness and transformation consultancy connecting assessment across six organizational pillars to prioritized gaps and a transformation roadmap.

Visit Artebello
Explore my work
Inside the system

Don’t just see the answer.
See how it gets there.

Explore the evidence, the tools, and the moments where a person stays in control.

Illustrative workspaceCedar Workshop

Fictional company. Curated data.
No live model or external actions.

  1. 01Question
  2. 02Retrieval
  3. 03Evidence
  4. 04Grounded answer
Explore this service
Guided simulation01 / 04

A useful question.

The assistant needs company policy, not a guess from a model’s training. Continue to inspect what it retrieves.

A simplified look inside the architecture. Every example is fictional; real systems require project-specific evaluation.

Start where you are.
Build what comes next.

One partner from the first assessment to the system your team depends on.

Assess.

Find the right opportunity. Understand the gaps. Leave with a clear implementation roadmap.

AI strategy & readiness

Build.

Connect knowledge, reasoning, and tools in a product that solves a real business problem.

Explore implementation services

Support.

Evaluate what happens in use. Improve reliability. Evolve the product as the business changes.

Discuss a project

AI is one layer.
I connect the rest.

Mobile apps, APIs, data, interfaces, and architecture. The engineering around the model is what turns a capability into a useful product.

Explore my expertise
Intelligent systemsAgents / RAG / MCP / LLM applications
Product engineeringFlutter / Next.js / NestJS / TypeScript
Architecture & directionData / Cloud / Product strategy / Leadership

Ideas worth
putting into practice.

Practical writing on building useful AI, informed by the work.

Consulting decisions

How to hire an AI consultant for a project that ships

A practical shortlist and proposal checklist for founders choosing an AI consultant: workflow scope, evidence, delivery ownership, and handover.

4 min readAgent reliability

How to evaluate a production AI agent before launch

Test an agent's tool use, approvals, retries, recovery, and task completion with a scenario matrix instead of judging only its final answer.

3 min readKnowledge systems

Production RAG starts with evidence, not embeddings

How to separate retrieval quality, access control, source provenance, and answer evaluation when building a useful company knowledge system.

3 min read
Read all insights
Engineer. Consultant. Founder.

I’m Paul. I like turning complicated problems into things people can actually use.

A little more about me Download my CV
Your next useful system

What could work
better?

Bring the business problem.
We’ll figure out the right next move.

Discuss a project