DANNY
PATON.

HOW I WORK / AI & ENGINEERING

I build the
workflow, too.

Part of my work is building software. Another part is designing how AI agents help build it.

I write and refine reusable skills: instructions that give an agent the context, steps and checks for a particular engineering task.

Get into the details

A WORKFLOW THAT KEEPS LEARNING

  1. 01
    Understand

    Read the task. Find the patterns.

  2. 02
    Build

    Make a focused change.

  3. 03
    Challenge

    Check the claims against the code.

  4. 04
    Verify

    Test the behavior that matters.

Carry the lesson into the next task.

WHAT’S AN AGENT SKILL?

A playbook for a specific job.

A skill describes when it applies, what context to read, how to approach the task and what a complete result looks like. I use these to make engineering practices available to an agent at the moment it needs them.

01When to use it02Context to read03Steps to follow04Evidence to return

THE PRACTICE

From instructions
to useful habits.

Open a practice to see how I approach it. The examples here are written for this portfolio.

01
CONTEXT & CONVENTIONS

Give the agent a map.

I write guidance that helps an agent find the existing components, types, design tokens and conventions before it starts changing code.

FOR EXAMPLE

A new results view should start with the shared table and data contract. The skill points the agent toward those patterns and asks it to check the callers before inventing another implementation.

A change that fits the product and its codebase.

02
REVIEW & VERIFICATION

Make review a real investigation.

I build review workflows around requirements, the current code and evidence. A convincing comment still needs to hold up when you follow the actual behavior.

FOR EXAMPLE

If a reviewer says an empty result will crash, trace that state through the caller and reproduce it. Fix the failure if it exists; record the evidence if the code already handles it; keep the question open if the evidence is incomplete.

Findings that explain what fails, why it matters and how it was checked.

03
FEEDBACK & MAINTENANCE

Teach the next run.

I turn recurring corrections into guidance that can be reused. That includes keeping agent instructions and review guidance aligned with the conventions they summarize.

FOR EXAMPLE

A keyboard-focus problem keeps appearing. Capture the recurring pattern, strengthen the relevant interaction guidance, and add a targeted check where it can catch the problem. Revisit that guidance when the component changes.

A lesson with a home, instead of the same comment on another pull request.

04
DELIVERY & HANDOFF

Finish with evidence.

I give workflows a clear finish: the relevant checks, the state of the change, and an explanation a reviewer can use. Local changes, published fixes and untested behavior need distinct labels.

FOR EXAMPLE

For an interface change, the handoff includes the behavior that changed, the checks run and browser observations at the relevant sizes. A later edit means revisiting any evidence that the edit could invalidate.

A reviewable result with clear limits and next steps.

PART OF THE DAY JOB

The tooling around
the product.

Alongside frontend delivery on the NASCAR platform, I work on reusable AI coding and review workflows. That sits within a broader team environment covering implementation, code review, build and test checks, pull requests, and technical documentation.

My contribution is writing and refining the practices I use, connecting guidance to real project conventions, and carrying review feedback forward. The wider repository includes shared skills contributed by other developers.

This page describes the approach. Employer skill files, internal prompts and review records stay private.

See the NASCAR platform

SO WHAT DOES THIS BRING TO A TEAM?

Software.
And a way to keep improving it.

Product engineering, practical AI tooling, and the ability to turn a useful working habit into a repeatable process.

Let’s talk