Recruiter-Facing AI Workflow Deck

# Recruiter-Facing AI Workflow Deck

## Slide 1: How I Use AI In Product Work

I use AI as an acceleration layer across research, design, engineering, QA, and documentation. The judgment stays human: what to build, what to trust, what to ship, and what to reject.

## Slide 2: The Operating Principle

AI is useful when it compresses the path from context to artifact.

It is risky when it replaces source-checking, product taste, security review, or user empathy.

## Slide 3: Research And Framing

How AI helps:

- Summarize product notes, tickets, calls, and source material.
- Identify gaps, assumptions, and edge cases.
- Compare possible directions.
- Draft briefs and sprint scopes.

Human review:

- Verify claims against source material.
- Decide what matters for the user and business.
- Remove generic or overconfident recommendations.

## Slide 4: Design And UX

How AI helps:

- Generate alternate flows.
- Critique hierarchy, copy, accessibility, and missing states.
- Explore interaction patterns.
- Pressure-test AI feature trust and fallback behavior.

Human review:

- Keep the design aligned with brand, context, and real user constraints.
- Decide the final hierarchy and interface language.
- Validate responsive behavior in the browser.

## Slide 5: Engineering

How AI helps:

- Inspect codebases.
- Draft implementation plans.
- Write scoped patches.
- Refactor repetitive code.
- Prepare PR summaries.

Human review:

- Read the generated code.
- Keep patterns aligned with the existing system.
- Avoid unnecessary abstractions.
- Run builds, tests, and manual QA.

## Slide 6: APIs And Integrations

How AI helps:

- Explore Postman collections and OpenAPI specs.
- Generate request examples.
- Find edge cases in auth, pagination, errors, and destructive actions.
- Review whether an API is clear enough for agent consumption.

Human review:

- Validate behavior against real requests.
- Protect secrets and sensitive data.
- Confirm side effects before automation.

## Slide 7: Content And Positioning

How AI helps:

- Draft copy variants.
- Audit portfolio content.
- Turn raw experience into clearer proof points.
- Keep tone consistent across pages.

Human review:

- Remove inflated language.
- Correct facts, dates, locations, and roles.
- Make the story specific to the person, product, or company.

## Slide 8: QA And Verification

How AI helps:

- Generate test scenarios.
- Review diffs for bug risk.
- Check missing states.
- Summarize deployment logs.

Human review:

- Run the app.
- Inspect the UI at real breakpoints.
- Confirm downloads, links, and forms.
- Decide whether residual risk is acceptable.

## Slide 9: Tools In The Workflow

Common tools:

- Claude Code for repo work and PR loops.
- ChatGPT and OpenAI for planning, critique, and alternative approaches.
- Cursor for daily AI-assisted coding.
- Figma and Figma AI for design exploration and handoff cleanup.
- Postman for API review and testing.
- Supabase for content and data-backed prototypes.
- Browser automation for local verification.

## Slide 10: What This Means For A Team

The value is not "AI generated this."

The value is faster cycles with better review discipline:

- Clearer plans.
- More complete states.
- Faster prototypes.
- Better QA coverage.
- More explicit handoffs.
- Cleaner PRs.

## Slide 11: Where I Draw The Line

I do not treat AI output as truth.

I do not ship code I have not reviewed.

I do not let AI make product judgment alone.

I do not expose secrets or sensitive data casually.

I do not optimize for novelty over usefulness.

## Slide 12: Summary

AI is now part of the product delivery stack. Used well, it lets a senior product builder move faster without lowering standards.

The point is not replacing craft. The point is spending more time on the parts where craft matters.