Knowledge Systems Architect
I design AI-ready knowledge systems
for messy real-world work.
I help organisations map their knowledge, structure their workflows, build useful AI systems, and connect people, documents, tools and agents into practical operating environments.
Move 37 thinking for Answer 42 problems.
02 · The system
A small public knowledge operating system.
Click any node to open it. Services connect to methods, methods to builds, builds to writing — and an agent layer makes all of it machine-readable.
⌘K to jump anywhere · Drag to pan · Scroll to zoom
03 · What I build
Chat is one interface. There are others — usually better.
01
Knowledge engines
Source-backed organisational memory with provenance, permissions and a clear update path.
02
Agentic workflows
Agents that propose drafts, decisions and next steps for human review — inside the work, not beside it.
03
Reporting layers
Specialist briefings for projects, risks, ops, sales, HR, ICT, compliance and finance — driven by the same knowledge base.
04 · How I work
Audit → Ingest → Assist → Report → Operate
01 · AUDIT
Audit.
Map the situation.
02 · INGEST
Ingest.
Build a structured knowledge base.
03 · ASSIST
Assist.
Deploy useful assistants.
04 · REPORT
Report.
Specialist reporting layers.
05 · OPERATE
Operate.
Make the work AI-operable.
05 · Work with me
Two service families, priced differently for a reason.
AI Systems · Knowledge
Strategic & technical AI work.
- AI Readiness Audit
- Knowledge Engine / RAG System
- Agentic Workflow Prototype
- AI Operations Retainer
- AI Strategy & Advisory
Web · Brand · Marketing
Practical production work.
- Website Builds & Updates
- Brand & Marketing Assets
- AI-Assisted Prototyping / Vibecoding
06 · Selected builds
From the lab.
Obsidian HKE Lite
A lightweight Holonic Knowledge Engine running over an Obsidian vault.
Open →AI Tutor / Lesson Planner
A drafting assistant for teachers, with a narrower source-backed student surface.
Open →Agent Capsule / Filesystem Triggers
Folder-native agents triggered by filesystem events — drop a file in, get a structured response back.
Open →07 · About
Luke Gleave Ijebor
Knowledge Systems Architect working on agentic AI, organisational knowledge systems, and practical tools that help people turn scattered information into usable intelligence. My work sits between systems thinking and practical build work.
More about me →08 · For agents
This site is designed to be read by machines, too.
Every page has structured metadata. The site exposes a manifest, a services file and project summaries — all consumable by an LLM agent at first contact.
{
"site": "LGI Knowledge Systems",
"endpoints": {
"welcome": "/agents/welcome",
"services": "/agents/services.json",
"projects": "/agents/projects.json"
},
"contact": "hello@example.com"
}09 · Get in touch
Let's talk about
your system.
Most engagements start with a short conversation about what's messy and what you'd like AI to actually do. I'll tell you whether it's an audit, a prototype or a website job.
Start a conversation