Paul Leung

Applied AI · Quality · Delivery

Evidence before adjectives.

I build AI systems, make their results visible, and help teams adopt them without losing human ownership.

See the evidence View the 10-slide story
16 years Quality, delivery and team leadership
3 years Hands-on applied AI building
10% → 80% Regression-scope expansion
16 MCP services In a self-built agent platform

Selected evidence

Built, transferred, and operated.

Each example proves something different. Work delivery and personal engineering are deliberately labelled rather than blended together.

Work · Active use 01

AI test automation

QA writes the intended workflow in natural-language test files. An Azure OpenAI worker executes bounded browser steps through Playwright MCP, collects evidence, and returns a result that can be reviewed.

Read the current case summary →
Regression scope 10% → 80% Approximate operational change
Proves measurable adoption and practical AI delivery Does not claim AI removed QA or guaranteed defect reduction
Work · Handoff 02

Agentic IT support PoC

Built a bot-first support journey, then coached the delivery team on prompts, tools, escalation and productionisation.

What this proves: I can shape a use case, prove the path, and transfer it instead of keeping the knowledge with one builder.
Personal · Engineering lab 03

Multi-agent platform

A self-built environment—Next.js, TypeScript, PostgreSQL, Kubernetes—for understanding orchestration, MCP tools, multi-model routing through LiteLLM, approvals, policy, audit and operations as one system.

Honest boundary: serious personal engineering, not presented as enterprise traffic, revenue or staffed production scale.
Personal · Live on this site 04

First-party analytics platform

A self-built Stats API measuring this site right now: public beacon with bot filtering and rate limits, token-authenticated product metrics, a k-anonymity guard in Postgres, and Metabase dashboards—deployed by GitOps on my own cluster.

Verify it live: loading this page sent one anonymous pageview—no cookies, no stored IPs, and visitor hashes that cannot be linked across days.

Why quality fits AI

Intent enters. Evidence comes out.

An AI can suggest, plan and act. It should not be the only party deciding that its own work is correct.

1

Human intent

State the work and expected outcome in language the domain team owns.

2

Bounded task

Define what the agent may do, where the step ends and what is forbidden.

3

Agent and tools

Give the model enough context and the smallest useful set of actions.

4

Evidence

Check system state, captured values, tool history or human acceptance.

5

Close or recover

Pass, retry, stop or escalate with a reason that another person can follow.

The prompt guides. Code still owns permissions and hard stops.

The agent acts. Evidence—not confidence—supports completion.

The team owns. AI does not erase accountability.

Read the complete product-neutral view How I understand and apply AI →

How I would scale

A core AI team, with shared ownership.

The central team should build a reusable road. It should not become the place where every other team throws its AI work.

01

Domain teams

Own the meaning

  • Workflow and exceptions
  • Domain tools and knowledge
  • Acceptance criteria
  • Final outcome sign-off
02

AI platform & enablement

Build the shared road

  • Agent foundation and tool standards
  • Evaluation and observability
  • Common controls and cost visibility
  • Coaching, patterns and adoption
03

SRE and Security

Remain independent gates

  • Reliability requirements
  • Security and data policy
  • Production readiness
  • Incident and audit standards

What I measure

Usage is a signal. Outcome is the result.

The management view is quality-adjusted productivity: useful delivery after including review, rework, defects, safety and cost.

Completion

Task success

Did the requested outcome actually happen?

Human cost

Review and rework

How much correction made the output usable?

Quality

Defect escape

Did speed move defects into a later stage?

Control

Safe execution

Were permissions and approval boundaries respected?

Flow

End-to-end time

Did the complete workflow become faster?

Economics

Cost per success

What did each accepted outcome actually cost?

About Paul

Quality leader turned applied AI builder.

My formal role is Scrum Master and Quality Assurance Manager. In practice, I work across AI solutioning, hands-on building, delivery leadership and team adoption.

Across 16 years in Hong Kong, I progressed from QA engineering into management, release delivery and technical team enablement. Over the last three years, applied AI became the natural continuation of that work.

I am strongest where an organisation needs someone who can prove the technical path, structure delivery, build team capability and move AI beyond a demonstration.

Applied AI Lead AI Platform Lead AI Adoption Lead Engineering Productivity

Continue the conversation

Build the evidence. Build the team. Earn the trust.

[email protected] CV / LinkedIn ↗ Presentation Hong Kong · Cantonese · Mandarin · English