feat: NLP patch insights + standalone demo mode

- dlib/ai/insights.py: pure-Python NLP analysis that correlates accepted
  AI suggestion operations/keywords/sections with submission outcomes
  (pending_review / published = positive, archived = negative)
- Backend: GET /api/v1/insights route + service + Pydantic schema
- Frontend: InsightsPanel component with bar charts for operation impact,
  section impact, and keyword signal lift scores
- Insights tab added to the version panel; compact preview on doc overview
- NEXT_PUBLIC_DEMO=true makes the webapp fully standalone: loads
  DEMO_DOCUMENTS / DEMO_SUBMISSIONS / DEMO_INSIGHTS from demo-data.ts,
  disables all mutating actions, shows a DEMO badge in the top bar
- apps/webapp/public/demo-cv.docx: static dummy CV (Alex Rivera) for demo
- scripts/gen_demo_cv.py: script to regenerate the demo DOCX
- .env.example: document NEXT_PUBLIC_DEMO flag

https://claude.ai/code/session_01LWxu2qrwY6BRjUFXXn7NiM
This commit is contained in:
Claude
2026-04-05 09:34:01 +00:00
parent 0f32d46404
commit 615d1bdb9e
12 changed files with 780 additions and 17 deletions

View File

@@ -0,0 +1,134 @@
'use client';
import type { InsightsResult } from '@/libs/api';
function Bar({ rate, positive }: { rate: number; positive?: boolean }) {
return (
<div style={{ flex: 1, height: 6, background: 'var(--border)', borderRadius: 3, overflow: 'hidden' }}>
<div style={{
width: `${Math.round(rate * 100)}%`,
height: '100%',
background: positive === false ? '#ef4444' : rate >= 0.6 ? '#22c55e' : rate >= 0.4 ? '#f59e0b' : '#94a3b8',
borderRadius: 3,
transition: 'width 0.3s',
}} />
</div>
);
}
function Pct({ v }: { v: number }) {
return <span style={{ fontVariantNumeric: 'tabular-nums', fontSize: 12, fontWeight: 600, color: v >= 0.6 ? '#16a34a' : v >= 0.4 ? '#d97706' : '#6b7280' }}>{Math.round(v * 100)}%</span>;
}
export default function InsightsPanel({ data }: { data: InsightsResult | null }) {
if (!data) return (
<div style={{ padding: '24px 0', color: 'var(--text-faint)', fontSize: 13, textAlign: 'center' }}>
Loading insights
</div>
);
if (!data.has_data) return (
<div style={{ padding: '24px 0', color: 'var(--text-faint)', fontSize: 13 }}>
Not enough data yet. Submit applications and mark outcomes to unlock insights.
</div>
);
return (
<div style={{ display: 'flex', flexDirection: 'column', gap: 20 }}>
{/* headline numbers */}
<div style={{ display: 'grid', gridTemplateColumns: 'repeat(3, 1fr)', gap: 8 }}>
{[
{ label: 'Total submissions', value: data.total_submissions },
{ label: 'Passed screening', value: data.positive_count },
{ label: 'Screening rate', value: `${Math.round(data.positive_rate * 100)}%` },
].map(({ label, value }) => (
<div key={label} style={{ border: '1px solid var(--border)', borderRadius: 6, padding: '8px 10px', background: 'var(--surface)' }}>
<div className="label" style={{ marginBottom: 3 }}>{label}</div>
<div style={{ fontSize: 18, fontWeight: 600 }}>{value}</div>
</div>
))}
</div>
{/* operation impact */}
{data.operation_impact.length > 0 && (
<section>
<div className="label" style={{ marginBottom: 8 }}>Patch operation impact</div>
<div style={{ display: 'flex', flexDirection: 'column', gap: 6 }}>
{data.operation_impact.map(op => (
<div key={op.operation} style={{ display: 'flex', alignItems: 'center', gap: 10 }}>
<span style={{ fontFamily: 'var(--font-mono)', fontSize: 11, width: 140, flexShrink: 0, color: 'var(--text-muted)' }}>
{op.operation}
</span>
<Bar rate={op.rate} />
<Pct v={op.rate} />
<span style={{ fontSize: 11, color: 'var(--text-faint)', width: 50, textAlign: 'right' }}>
{op.positive}/{op.total}
</span>
</div>
))}
</div>
<p style={{ fontSize: 11, color: 'var(--text-faint)', marginTop: 6 }}>
% of accepted patches of this type in submissions that passed screening.
</p>
</section>
)}
{/* section impact */}
{data.section_impact.length > 0 && (
<section>
<div className="label" style={{ marginBottom: 8 }}>CV section impact</div>
<div style={{ display: 'flex', flexDirection: 'column', gap: 6 }}>
{data.section_impact.map(s => (
<div key={s.section} style={{ display: 'flex', alignItems: 'center', gap: 10 }}>
<span style={{ fontFamily: 'var(--font-mono)', fontSize: 11, width: 80, flexShrink: 0, color: 'var(--text-muted)' }}>
{s.section}
</span>
<Bar rate={s.positive_rate} />
<Pct v={s.positive_rate} />
<span style={{ fontSize: 11, color: 'var(--text-faint)', width: 50, textAlign: 'right' }}>
{s.count} edits
</span>
</div>
))}
</div>
</section>
)}
{/* keyword signals */}
{(data.top_positive_keywords.length > 0 || data.top_negative_keywords.length > 0) && (
<section>
<div className="label" style={{ marginBottom: 8 }}>Keyword signals</div>
<div style={{ display: 'grid', gridTemplateColumns: '1fr 1fr', gap: 12 }}>
<div>
<div style={{ fontSize: 11, color: '#16a34a', fontWeight: 600, marginBottom: 6 }}>Positive signals</div>
<div style={{ display: 'flex', flexDirection: 'column', gap: 4 }}>
{data.top_positive_keywords.map(k => (
<div key={k.keyword} style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center' }}>
<span style={{ fontSize: 12, fontFamily: 'var(--font-mono)', color: 'var(--text)' }}>{k.keyword}</span>
<span style={{ fontSize: 11, color: '#16a34a' }}>+{k.positive_count} ({k.lift}×)</span>
</div>
))}
</div>
</div>
<div>
<div style={{ fontSize: 11, color: '#dc2626', fontWeight: 600, marginBottom: 6 }}>Negative signals</div>
<div style={{ display: 'flex', flexDirection: 'column', gap: 4 }}>
{data.top_negative_keywords.length === 0
? <span style={{ fontSize: 12, color: 'var(--text-faint)' }}>None yet</span>
: data.top_negative_keywords.map(k => (
<div key={k.keyword} style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center' }}>
<span style={{ fontSize: 12, fontFamily: 'var(--font-mono)', color: 'var(--text)' }}>{k.keyword}</span>
<span style={{ fontSize: 11, color: '#dc2626' }}>{k.negative_count}×</span>
</div>
))}
</div>
</div>
</div>
<p style={{ fontSize: 11, color: 'var(--text-faint)', marginTop: 6 }}>
Keywords extracted from accepted AI suggestions, split by outcome.
</p>
</section>
)}
</div>
);
}