Human-Centered AI Systems Designer

Human-centered AIsystems forreal operationalwork.

I help teams simplify communication, workflows, and decision-making through responsible AI, design, and systems thinking.

Turning complexity into clear, useful systems.

I simplify complex operations into clearer tools, language, and AI-supported flows.

Visual Clarity

Communication Systems

Structure that makes complex ideas easier to see, explain, and use.

Human AI

AI Workflow Design

Drafting, sorting, and decision support with human review built in.

Calm Operations

Automation & Productivity Systems

Lean operating spaces for tasks, notes, documents, and handoffs.

Teams and organizations dealing with operational overwhelm.

The best fit is not people chasing AI trends. It is teams dealing with communication overload, repeated work, and unclear systems.

Operations-heavy teamsHigh-volume requests, handoffs, and documentation.
Public service and frontline service environmentsReal people, policy pressure, and service complexity.
Small businesses and foundersToo many tools, decisions, and unfinished systems.
Communication-heavy teamsToo much rewriting, follow-up, and explanation.
Responsible AI explorersAI adoption without losing control or trust.

A practical process for systems people can actually use.

Start with the work, find the friction, then introduce AI only where it has a clear purpose.

Understand the workflowFind the actual job, not the assumed tool.
Identify frictionSpot drag, handoffs, missing context, and repeated decisions.
Simplify the systemDefine inputs, outputs, owners, and rules.
Design the AI-supported flowBuild templates, prompts, review points, and handoffs.
Keep humans in reviewKeep approval, judgment, and accountability visible.
Test and improveUse real pressure to improve the system.

Proof, not promises.

Each project needs visible artifacts: maps, interfaces, prompts, outputs, and before/after states.

Before and after workflow thumbnail from the MainPageImages folder.
Before → After WorkflowFriction becomes a path.
Input, AI support, human review, and final output thumbnail from the MainPageImages folder.
AI + Human Review FlowAI stays reviewable.
System diagrams thumbnail from the MainPageImages folder.
System DiagramsRoles and logic.
Prompt architecture thumbnail from the MainPageImages folder.
Prompt ArchitectureRules and guardrails.
Process maps thumbnail from the MainPageImages folder.
Process MapsReusable operations.

Built from real operational environments, not ideal demos.

The memorable model: pressure becomes structure, structure becomes support, and support stays human-confirmed.

Responsible AI should make people more capable.

Trust is the constraint. AI only helps when people can verify it.

AI should support humans, not replace themAutomation supports structure, not accountability.
Clarity over complexityMake the next decision easier to see.
Human review mattersEvery output needs a checkpoint.
Privacy and responsibility come firstProtect sensitive context and avoid invented facts.
Systems should reduce stressGood systems lower cognitive load.

Work that turns messy moments into clear systems.

Four projects. Each turns pressure into a visible operating path.

Technology should not make people feel smaller. Good systems reduce confusion, protect judgment, and make work easier to understand.

Fabian Delven designs systems that feel calm enough to use.

I work between design, automation, and AI implementation. The goal is not a perfect demo. The goal is calmer decisions for real people doing real work.

More About Me

Have messy work? Let's make it clear.

Bring the rough version: scattered notes, repeated steps, fuzzy decisions, or the tool stack that almost works.