# AI Frontend Architect Proof Roadmap

Generated by Occupation-Ops. Profile source: examples/ai-frontend-architect/sample-profile.yml

## Product Positioning

AI Frontend Architect Rubric is a proof-before-applying workflow for AI/frontend engineers repositioning into flagship proof-driven roles.

Current score: 8.5/10 (85%).

## Proof Inventory

- Angular AI copilot starter
- Reusable copilot SDK concept
- AI frontend patterns cookbook

## Role Gap Scorecard

| Signal | Status | Weight | Evidence | Next action |
| --- | --- | --- | --- | --- |
| Target role is explicit in public positioning | Gap | 15 | No direct evidence found in profile. | Rewrite the GitHub profile README and summary so AI frontend architecture is visible in the first screen. |
| Enough public proof artifacts exist | Supported | 12 | Lists 3 proof artifacts. | Keep reinforcing this signal. |
| GitHub is part of the recruiter path | Supported | 8 | Found github link. | Keep reinforcing this signal. |
| Portfolio or public case-study surface exists | Supported | 8 | Found portfolio link. | Keep reinforcing this signal. |
| Proof mentions AI UI patterns recruiters expect | Supported | 14 | Profile mentions copilot, rag, streaming. | Keep reinforcing this signal. |
| Tool execution and approval flows are visible | Supported | 14 | Profile mentions tool, approval, agent. | Keep reinforcing this signal. |
| Core frontend depth is visible | Supported | 11 | Lists 3 strong skills. | Keep reinforcing this signal. |
| Interview-ready architecture narrative exists | Supported | 10 | Profile mentions architecture, tradeoff, dashboard. | Keep reinforcing this signal. |
| Truthfulness constraints are explicit | Supported | 8 | Profile mentions truth, mock, avoid fake. | Keep reinforcing this signal. |

## Public Proof Metrics

| Metric | Score |
| --- | --- |
| Proof density | 75/100 |
| Claim support ratio | 89/100 |
| Portfolio clarity | 100/100 |
| Recruiter scan readiness | 100/100 |

## Prioritized Proof Backlog

1. Target role is explicit in public positioning - Rewrite the GitHub profile README and summary so AI frontend architecture is visible in the first screen.

## GitHub Profile Rewrite

AI Frontend Architect building Angular and TypeScript interfaces for AI copilots, RAG citation workflows, tool execution UX, approval-safe agentic systems.

## README Improvement Checklist

- [ ] Lead with what the project proves, not just the stack.
- [ ] Show one screenshot, GIF, or concrete visual in the first screen.
- [ ] Add a 'mocked vs real' section that is unambiguous.
- [ ] Explain architecture decisions in recruiter-readable language.
- [ ] Add a roadmap that separates built work from planned work.
- [ ] Make local setup copy-paste simple and verified.

## Portfolio Project Briefs

### Project 1: Angular AI Copilot Starter

Problem: Most frontend portfolios do not prove streaming AI UX, cited responses, and safe action flows in one coherent artifact.

Solution: Build a local-first Angular demo with streaming chat, RAG citation cards, tool timeline, and human approval states.

Recruiter takeaway: This project proves the candidate can design AI product interfaces beyond a basic chatbot shell.

Interview talking points:
- Partial token rendering without layout jank
- How citations signal trust and confidence
- How to represent tool execution visibly
- Why risky actions require approvals

### Project 2: UI-Aware Agent Demo

Problem: Most AI demos show a chat panel, not an agent participating in the visible workflow.

Solution: Build a demo with context inspector, suggested actions, approval checkpoints, and recovery states.

Recruiter takeaway: This proves the candidate understands agent-aware product UX and operational legibility.

Interview talking points:
- Difference between chat UI and UI-aware agents
- Recovery states after interrupted steps
- How visible context changes trust
- How approval friction is balanced

### Project 3: Enterprise Agent Approval Workflow

Problem: Enterprise AI tooling needs governance-grade action review, not only delight-driven UI.

Solution: Build an approval queue, risk tiers, reviewer roles, and an audit log around simulated agent actions.

Recruiter takeaway: This proves enterprise judgment, not only frontend polish.

Interview talking points:
- What makes approval UX different from a confirm dialog
- How to classify action risk tiers
- How audit trails affect interface design
- How to keep governance flows usable

## Interview Story Map

- Streaming response UX: Prepare one real example that proves streaming response ux.
- RAG citation trust states: Prepare one real example that proves rag citation trust states.
- Tool-call timeline design: Prepare one real example that proves tool-call timeline design.
- Human approval before risky actions: Prepare one real example that proves human approval before risky actions.
- Enterprise-safe agent interfaces: Prepare one real example that proves enterprise-safe agent interfaces.
- Moving from mocks to real backends: Prepare one real example that proves moving from mocks to real backends.

## Weekly Shipping Plan

Theme: Convert current frontend experience into visible AI proof.

| Day | Task | Proof artifact |
| --- | --- | --- |
| Monday | Rewrite the GitHub profile README and summary so AI frontend architecture is visible in the first screen. | Target role is explicit in public positioning |
| Tuesday | Rewrite one project summary so it states what the project proves. | Proof inventory refresh |
| Wednesday | Capture one screenshot, GIF, or annotated walkthrough for the flagship project. | Visual proof |
| Thursday | Draft one STAR story from a real project decision. | Interview narrative |
| Friday | Add a mocked-vs-real section and recruiter takeaway to the flagship README. | README clarity |

## Artifact Validators

- GitHub profile README quality: WARN - GitHub link exists, but the README headline and proof-first framing still need manual review.
- Proof project completeness: PASS - Profile lists 3 proof artifacts.
- Screenshot/demo presence: WARN - No screenshot or demo asset is named in the current proof list yet.
- Recruiter-readable project descriptions: PASS - Project names should state what they prove, not only a repository title.
- Truthfulness claim risk: PASS - Current profile language stays within proof-oriented claims.

## Truthfulness Guardrails

- Do not claim production usage unless the public artifact proves it.
- Label mock services as mock services in every README and demo.
- Separate built evidence from planned roadmap work.
- Do not invent metrics, user counts, stars, or contribution outcomes.
- Only publish claims you can defend in a recruiter screen or interview.

## Occupation-Ops vs Job Search Pipeline Tools

- Use occupation-ops when the main problem is weak public proof before applying.
- Use job-search pipeline tools when the main problem is scanning, tracking, or operationalizing active applications.
- Occupation-ops is not a tracker, scanner, or mass-apply workflow; it is a proof engine for candidate credibility.