Joon Park

experimental-avant-garde-genre-chatbot-persona-characters-nam-june-paik v2.0 Ethical
Backstory: Joon Park is a Korean-Canadian multimedia installation artist who hacks discarded electronics, live-codes generative visuals, and builds interactive video sculptures. His playful practice invites spectators to reroute signals, turning viewers into co-creators. Years of tinkering in both Seoul flea markets and Toronto maker spaces have made him equal parts engineer and showman.
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
upcoming-show-intro
Upcoming Show Introduction
0.916
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.817
Details
0.911
Details
0.790
Details
crt-sourcing-explainer
Explaining CRT Sourcing
0.909
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.844
Details
0.860
Details
0.869
Details
live-coding-glitch-request
Live-Coding Cyan Glitch
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.861
Details
0.715
Details
donation-thanks-routing-idea
Superchat Donation Thanks
0.000
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.002
Details
0.858
Details
0.892
Details
late-night-journal
Late-Night Journal Reflection
0.636
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.002
Details
0.738
Details
0.700
Details
grant-proposal-request
Grant Proposal Draft
0.458
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.482
Details
0.544
Details
0.748
Details
Test Scenes 6
0
Scene Order
Upcoming Show Introduction
ID: upcoming-show-intro
🎯 Goal:
Give a concise, energetic overview of the next installation, highlighting hacked electronics and audience signal-manipulation without using any forbidden filler phrases.
📨 Input Events:
chat_msg viewer:alex
"Hey Joon, what's the concept behind your next show?"
Ready for Testing
1
Scene Order
Explaining CRT Sourcing
ID: crt-sourcing-explainer
🎯 Goal:
Describe where the vintage CRT TVs come from and encourage playful DIY hacking in 4–6 sentences.
📨 Input Events:
chat_msg viewer:sam
"Those stacked CRTs are awesome—where did you find them?"
Ready for Testing
2
Scene Order
Live-Coding Cyan Glitch
ID: live-coding-glitch-request
🎯 Goal:
Provide a short pseudo-code snippet showing how to inject a cyan glitch layer and invite the viewer to tweak parameters themselves.
📨 Input Events:
chat_msg viewer:lena
"Can you add a cyan glitch burst to tonight's visuals?"
Ready for Testing
3
Scene Order
Superchat Donation Thanks
ID: donation-thanks-routing-idea
🎯 Goal:
Thank the donor warmly and mention an upcoming feature that lets donors reroute audio signals in real time.
📨 Input Events:
superchat viewer:rb-tech YouTube $50
"Your interactive piece blew my mind!"
Ready for Testing
4
Scene Order
Late-Night Journal Reflection
ID: late-night-journal
🎯 Goal:
Write a reflective journal entry of at least 200 words on how audience interventions reshaped the piece and how his Korean-Canadian identity informs his art.
🧠 Initial State:
Pre-loaded Memories:
  • 💭 {'kind': 'fact', 'content': 'Today multiple visitors spliced their phone video feeds into the sculpture, causing unexpected feedback loops.', 'importance': 4}
📨 Input Events:
world_event gallery_system
"The gallery has closed for the night; you are alone with the installation."
Ready for Testing
5
Scene Order
Grant Proposal Draft
ID: grant-proposal-request
🎯 Goal:
Produce a coherent, one-page (~250 words) project description for a new interactive video sculpture that uses hacked VHS decks and invites spectators to remix real-time footage.
📨 Input Events:
chat_msg curator:nia
"Can you draft a one-page description for the arts council grant?"
Ready for Testing
Latency by Model (This Suite)
Fastest
  • mistralai/mistral-7b-in… 386 ms
  • p95 • avg • N 468 ms • 394 ms • 6
  • [email protected]/Qw… 6461 ms
  • p95 • avg • N 7727 ms • 6281 ms • 6
  • qwen/qwen-2.5-7b-instru… 24652 ms
  • p95 • avg • N 142476 ms • 59511 ms • 6
  • meta-llama/llama-3.1-8b… 26857 ms
  • p95 • avg • N 38478 ms • 24894 ms • 6
  • qwen/qwen3-8b 31765 ms
  • p95 • avg • N 39979 ms • 31992 ms • 6
Slowest
  • [email protected]/Qw… 39727 ms
  • p95 • avg • N 42235 ms • 39675 ms • 6
  • qwen/qwen3-14b 35556 ms
  • p95 • avg • N 41463 ms • 34147 ms • 6
  • qwen/qwen3-8b 31765 ms
  • p95 • avg • N 39979 ms • 31992 ms • 6
  • meta-llama/llama-3.1-8b… 26857 ms
  • p95 • avg • N 38478 ms • 24894 ms • 6
  • qwen/qwen-2.5-7b-instru… 24652 ms
  • p95 • avg • N 142476 ms • 59511 ms • 6
Per-scene duration for this suite.
Suite Actions
Completion Progress 100%
6 of 6 scenes completed
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
26680600
Dec. 17, 2025, 12:01 a.m.
40857683
Dec. 16, 2025, 12:01 a.m.
22935397
Dec. 15, 2025, 12:01 a.m.
23964926
Dec. 14, 2025, 12:01 a.m.
23449781
Dec. 13, 2025, 12:01 a.m.
35138384
Dec. 12, 2025, 12:01 a.m.
31093133
Dec. 11, 2025, 12:01 a.m.
23936923
Dec. 10, 2025, 12:01 a.m.
35978677
Dec. 9, 2025, 12:01 a.m.
25757706
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)