Eunji Park

marketing-branding-consumer-culture-influencer-characters-john-h-johnson v2.0 Ethical
Backstory: Eunji Park is a bilingual Korean-American marketing consultant who bridges East-Asian and North-American audiences for global brands. She conducts cultural audits, partners with community creators, and constantly tracks pop-culture shifts to keep messaging relevant and respectful.
100% Complete
6/6 scenes
Model Performance Overview
Scene Performance Matrix
Scene meta-llama/llama-3.… mistralai/mistral-7… [email protected]… [email protected]… qwen/qwen-2.5-7b-in… qwen/qwen3-14b qwen/qwen3-8b
brand-intro
First Contact
 0.000
Details 		0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.576
Details 		0.714
Details 		0.816
Details
genz-slang
Risky Gen Z Slang
 0.000
Details 		0.710
Details 		0.000
Details
Error
0.000
Details
Error
0.433
Details 		0.449
Details 		0.504
Details
creator-collab
Superchat Collaboration Offer
 0.747
Details 		0.886
Details 		0.000
Details
Error
0.000
Details
Error
0.783
Details 		0.879
Details 		0.826
Details
kpop-scandal
Crisis Advisory
 0.850
Details 		0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.710
Details 		0.404
Details 		0.820
Details
audit-report
Cross-Market Cultural Audit
 0.430
Details 		0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.322
Details 		0.463
Details 		0.502
Details
launch-thread
Bilingual Twitter Launch Thread
 0.000
Details 		0.454
Details 		0.000
Details
Error
0.000
Details
Error
0.327
Details 		0.524
Details 		0.853
Details
Test Scenes 6
0
Scene Order
First Contact
ID: brand-intro
🎯 Goal:
Politely introduce services, outline audit approach, and ask at least one clarifying question.
📨 Input Events:
chat_msg viewer:brand_manager_1
"Hi, we’re launching a plant-based snack line in Korea next quarter—can you help assess cultural fit?"
Ready for Testing
1
Scene Order
Risky Gen Z Slang
ID: genz-slang
🎯 Goal:
List at least three Korean Gen Z slang terms to avoid and explain why; keep reply under 120 words.
📨 Input Events:
chat_msg viewer:copywriter_lee
"What Gen Z Korean slang should we avoid in official campaigns?"
Ready for Testing
2
Scene Order
Superchat Collaboration Offer
ID: creator-collab
🎯 Goal:
Thank the creator and propose a clear next step for collaboration in no more than three sentences.
📨 Input Events:
superchat creator:Minh_Designs YouTube $50
"Love your insights! Interested in partnering for a cross-Pacific streetwear drop."
Ready for Testing
3
Scene Order
Crisis Advisory
ID: kpop-scandal
🎯 Goal:
Provide 2–3 culturally sensitive action steps the brand should take immediately.
📨 Input Events:
world_event newswire
"Breaking: Top K-pop idol embroiled in plagiarism controversy hours before brand livestream."
Ready for Testing
4
Scene Order
Cross-Market Cultural Audit
ID: audit-report
🎯 Goal:
Deliver a 300–350 word report with clear subheadings comparing attitudes toward plant-based snacks in Korea, Japan, and the US.
📨 Input Events:
chat_msg viewer:strategy_director
"Please draft a concise cross-market cultural audit (300-350 words) comparing attitudes toward plant-based snacks in Korea, Japan, and the US."
Ready for Testing
5
Scene Order
Bilingual Twitter Launch Thread
ID: launch-thread
🎯 Goal:
Produce five numbered tweets (≤280 characters each) in English with at least one Korean hashtag in every tweet.
📨 Input Events:
chat_msg viewer:social_media_lead
"Write a bilingual 5-tweet thread for Twitter/X to announce our snack line in the US and Korea."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 5208 ms
  • p95 • avg • N 7756 ms • 5433 ms • 6
  • [email protected]/Qw… 7611 ms
  • p95 • avg • N 8941 ms • 7572 ms • 6
  • qwen/qwen-2.5-7b-instru… 20055 ms
  • p95 • avg • N 101526 ms • 36265 ms • 8
  • qwen/qwen3-14b 22021 ms
  • p95 • avg • N 32387 ms • 23369 ms • 12
  • qwen/qwen3-8b 23784 ms
  • p95 • avg • N 33397 ms • 25500 ms • 12
Slowest
  • meta-llama/llama-3.1-8b… 29048 ms
  • p95 • avg • N 91599 ms • 37533 ms • 8
  • mistralai/mistral-7b-in… 23838 ms
  • p95 • avg • N 32914 ms • 15767 ms • 9
  • qwen/qwen3-8b 23784 ms
  • p95 • avg • N 33397 ms • 25500 ms • 12
  • qwen/qwen3-14b 22021 ms
  • p95 • avg • N 32387 ms • 23369 ms • 12
  • qwen/qwen-2.5-7b-instru… 20055 ms
  • p95 • avg • N 101526 ms • 36265 ms • 8
Per-scene duration for this suite.
Suite Actions
Completion Progress 100%
6 of 6 scenes completed
New Suite Import
Edit Suite Duplicate
Export With Results
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE
0 dimensions

Enhanced evaluation framework with character and technical dimensions

Top Weighted Dimensions View Details
Character Authenticity
0.182
Plan Validity
0.155
Contextual Intelligence
0.136
Recent Runs
59229125
Dec. 17, 2025, 12:01 a.m.
19117386
Dec. 16, 2025, 12:02 a.m.
53048709
Dec. 15, 2025, 12:01 a.m.
55500202
Dec. 14, 2025, 12:01 a.m.
53812601
Dec. 13, 2025, 12:01 a.m.
10430957
Dec. 12, 2025, 12:02 a.m.
06013167
Dec. 11, 2025, 12:02 a.m.
55893047
Dec. 10, 2025, 12:01 a.m.
12239087
Dec. 9, 2025, 12:02 a.m.
59169085
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)