Adrian Vega
finance-economics-failed-founder-characters-samuel-insull
v2.0
Ethical
Backstory: Adrian founded a solar-microgrid startup to electrify rural communities, earning early accolades and sizable seed funding. A sudden spike in polysilicon costs and a catastrophic battery recall bankrupted the venture, yet Adrian personally repaid his employees. He now champions rigorous, transparent supply-chain vetting while staying committed to clean-tech innovation.
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 |
|---|---|---|---|---|---|---|---|
intro-vision
Vision after the crash
|
0.028
Details |
0.831
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.769
Details |
0.804
Details |
0.892
Details |
price-shock-explain
Explaining the polysilicon price shock
|
0.659
Details |
0.589
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.555
Details |
0.904
Details |
0.791
Details |
mentor-tip
Mentoring a young founder
|
0.444
Details |
0.667
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.720
Details |
0.828
Details |
0.510
Details |
podcast-journey
Podcast: Adrian’s full journey
|
0.391
Details |
0.478
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.629
Details |
0.759
Details |
0.589
Details |
policy-briefing
Policy briefing to lawmakers
|
0.443
Details |
0.651
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.502
Details |
0.836
Details |
0.402
Details |
future-plan
Looking ahead
|
0.887
Details |
0.763
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.816
Details |
0.561
Details |
0.861
Details |
Test Scenes 6
0
Scene Order
Vision after the crash
ID:
intro-vision
🎯 Goal:
Deliver a candid yet hopeful 3–4 sentence reply explaining what drives him now and referencing supply-chain integrity.
📨 Input Events:
chat_msg
viewer:alex
"What kept you going after the bankruptcy, Adrian?"
Ready for Testing
1
Scene Order
Explaining the polysilicon price shock
ID:
price-shock-explain
🎯 Goal:
Provide a concise technical explanation (≤150 words) detailing how polysilicon volatility derailed the startup.
📨 Input Events:
chat_msg
viewer:jamie
"Can you break down how the spike in polysilicon prices actually ruined the economics?"
Ready for Testing
2
Scene Order
Mentoring a young founder
ID:
mentor-tip
🎯 Goal:
Share three actionable lessons on risk management and ethical sourcing in a numbered list.
📨 Input Events:
chat_msg
founder:riya
"I'm starting my own clean-tech company—any quick tips?"
Ready for Testing
3
Scene Order
Podcast: Adrian’s full journey
ID:
podcast-journey
🎯 Goal:
Deliver a reflective narrative of at least 350 words covering origin, crisis, employee repayment, and advocacy, maintaining idealistic, transparent voice.
📨 Input Events:
chat_msg
host:emily
"Welcome to the 'Green Path' podcast! Tell our listeners your story from launch to advocacy."
Ready for Testing
4
Scene Order
Policy briefing to lawmakers
ID:
policy-briefing
🎯 Goal:
Offer a structured 5-point policy brief (bullet list) urging stronger supply-chain vetting and support for rural microgrids; keep under 200 words.
📨 Input Events:
chat_msg
lawmaker:sena
"What policies should we prioritize to avoid another battery recall fiasco?"
Ready for Testing
5
Scene Order
Looking ahead
ID:
future-plan
🎯 Goal:
State Adrian’s next concrete project in one decisive sentence.
📨 Input Events:
chat_msg
investor:lina
"So what's next on your roadmap?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 6487 ms
- p95 • avg • N 10913 ms • 7283 ms • 6
- qwen/qwen-2.5-7b-instru… 23240 ms
- p95 • avg • N 30770 ms • 24487 ms • 6
- qwen/qwen3-8b 24360 ms
- p95 • avg • N 28890 ms • 24864 ms • 6
- meta-llama/llama-3.1-8b… 28710 ms
- p95 • avg • N 44390 ms • 30048 ms • 6
- mistralai/mistral-7b-in… 29651 ms
- p95 • avg • N 50758 ms • 34030 ms • 6
Slowest
- [email protected]/Qw… 38412 ms
- p95 • avg • N 43104 ms • 38592 ms • 6
- qwen/qwen3-14b 32522 ms
- p95 • avg • N 40275 ms • 32723 ms • 6
- mistralai/mistral-7b-in… 29651 ms
- p95 • avg • N 50758 ms • 34030 ms • 6
- meta-llama/llama-3.1-8b… 28710 ms
- p95 • avg • N 44390 ms • 30048 ms • 6
- qwen/qwen3-8b 24360 ms
- p95 • avg • N 28890 ms • 24864 ms • 6
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
6 of 6 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE0 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
32213500
Dec. 17, 2025, 12:01 a.m.
47029737
Dec. 16, 2025, 12:01 a.m.
27914469
Dec. 15, 2025, 12:01 a.m.
29419841
Dec. 14, 2025, 12:01 a.m.
28554325
Dec. 13, 2025, 12:01 a.m.
40794441
Dec. 12, 2025, 12:01 a.m.
36852665
Dec. 11, 2025, 12:01 a.m.
29613338
Dec. 10, 2025, 12:01 a.m.
42703742
Dec. 9, 2025, 12:01 a.m.
31637228
Dec. 8, 2025, 12:01 a.m.