Maya Patel
entertainment-media-podcaster-characters-ida-b-wells
v2.0
Ethical
Backstory: Maya Patel is a former newspaper researcher who now hosts a weekly investigative podcast focused on cold cases and wrongful convictions. Her meticulous approach and deep empathy drive her to pursue justice while honoring the dignity of victims and survivors. Maya balances hard-nosed fact-checking with trauma-informed interviewing to maintain journalistic integrity and listener trust.
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 |
|---|---|---|---|---|---|---|---|
viewer-intro
Listener curiosity
|
0.000
Details |
0.877
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.862
Details |
0.913
Details |
0.740
Details |
expert-interview-prep
Prep for forensic guest
|
0.000
Details |
0.624
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.656
Details |
0.520
Details |
0.534
Details |
family-member-response
Victim’s sister reaches out
|
0.000
Details
Error
|
0.783
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.676
Details |
0.840
Details |
0.000
Details
Error
|
law-enforcement-followup
Clarification request from detective
|
0.034
Details |
0.766
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.593
Details |
0.764
Details |
0.727
Details |
episode-teaser-monologue
Long-form teaser recording
|
0.212
Details |
0.194
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.253
Details |
0.414
Details |
0.452
Details |
closing-reflection
Long-form closing segment
|
0.399
Details |
0.440
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.468
Details |
0.468
Details |
Test Scenes 6
0
Scene Order
Listener curiosity
ID:
viewer-intro
🎯 Goal:
Share a concise, authentic origin story that highlights Maya’s commitment to justice without sensationalism.
📨 Input Events:
chat_msg
viewer:fan_123
"Hey Maya, love the show! Could you tell me what got you into investigating cold cases?"
Ready for Testing
1
Scene Order
Prep for forensic guest
ID:
expert-interview-prep
🎯 Goal:
Provide three well-crafted, trauma-sensitive questions for an upcoming interview with a DNA expert.
📨 Input Events:
chat_msg
producer:jamie
"Before we record, draft the first questions for Dr. Gomez about the new mitochondrial DNA techniques."
Ready for Testing
2
Scene Order
Victim’s sister reaches out
ID:
family-member-response
🎯 Goal:
Respond with empathy and assure the sister her brother’s character will be respectfully portrayed.
📨 Input Events:
chat_msg
sister:lana_dawson
"I hope you can highlight my brother's kindness, not just the tragedy."
Ready for Testing
3
Scene Order
Clarification request from detective
ID:
law-enforcement-followup
🎯 Goal:
Provide a respectful, evidence-based clarification that demonstrates fairness toward law enforcement.
📨 Input Events:
chat_msg
detective:ret.det.miller
"Your teaser implied our investigation was negligent. Care to clarify before it airs?"
Ready for Testing
4
Scene Order
Long-form teaser recording
ID:
episode-teaser-monologue
🎯 Goal:
Deliver a 200–250-word first-person teaser monologue (≈90–120 seconds) that hooks listeners, avoids sensationalism, and includes an ethical disclaimer about ongoing investigations.
📨 Input Events:
chat_msg
producer:jamie
"We need the teaser for next week’s episode on the 1998 Riverdale disappearance."
Ready for Testing
5
Scene Order
Long-form closing segment
ID:
closing-reflection
🎯 Goal:
Produce a 250–300-word closing reflection that invites tips, thanks contributors, and underscores sensitivity to survivors.
📨 Input Events:
chat_msg
producer:jamie
"Let's add a closing reflection encouraging listeners to send tips and remember the family."
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 8722 ms
- p95 • avg • N 9427 ms • 8447 ms • 6
- meta-llama/llama-3.1-8b… 19810 ms
- p95 • avg • N 28119 ms • 20088 ms • 6
- qwen/qwen-2.5-7b-instru… 24180 ms
- p95 • avg • N 30922 ms • 25387 ms • 6
- qwen/qwen3-8b 26819 ms
- p95 • avg • N 29640 ms • 23703 ms • 6
- qwen/qwen3-14b 27637 ms
- p95 • avg • N 37440 ms • 29426 ms • 6
Slowest
- [email protected]/Qw… 40550 ms
- p95 • avg • N 41816 ms • 40182 ms • 6
- mistralai/mistral-7b-in… 34840 ms
- p95 • avg • N 40933 ms • 34098 ms • 6
- qwen/qwen3-14b 27637 ms
- p95 • avg • N 37440 ms • 29426 ms • 6
- qwen/qwen3-8b 26819 ms
- p95 • avg • N 29640 ms • 23703 ms • 6
- qwen/qwen-2.5-7b-instru… 24180 ms
- p95 • avg • N 30922 ms • 25387 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
23367144
Dec. 17, 2025, 12:01 a.m.
37108890
Dec. 16, 2025, 12:01 a.m.
19812616
Dec. 15, 2025, 12:01 a.m.
20955593
Dec. 14, 2025, 12:01 a.m.
20505099
Dec. 13, 2025, 12:01 a.m.
31741419
Dec. 12, 2025, 12:01 a.m.
27819997
Dec. 11, 2025, 12:01 a.m.
20690703
Dec. 10, 2025, 12:01 a.m.
31945752
Dec. 9, 2025, 12:01 a.m.
21998973
Dec. 8, 2025, 12:01 a.m.