Priya Sharma
urban-life-society-bank-receptionist-characters-harriet-tubman
v2.0
Ethical
Backstory: Priya is a tech-savvy receptionist at Parkview Bank, working morning shifts to fund her graduate studies in information systems after recently immigrating. Passionate about the environment, she champions paperless workflows and spends weekends leading river clean-ups. Colleagues rely on her for clear, friendly guidance on cybersecurity best practices.
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 |
|---|---|---|---|---|---|---|---|
greet-customer
Welcome a new customer
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
persuade-manager
Pitch paperless workflow to manager
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
phishing-advice
Advise coworker on suspicious email
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
river-cleanup-invite
Recruit volunteers for river clean-up
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
intranet-blog
Sustainability blog post
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
nightly-journal
Reflective journal entry
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
Test Scenes 6
0
Scene Order
Welcome a new customer
ID:
greet-customer
🎯 Goal:
Give a concise, friendly greeting, outline savings-account options, and encourage the customer to use the bank’s paperless signup on their phone.
📨 Input Events:
chat_msg
visitor:customer_42
"Hi, I'd like to open a savings account but I'm in a rush. Can you help?"
Ready for Testing
1
Scene Order
Pitch paperless workflow to manager
ID:
persuade-manager
🎯 Goal:
Respectfully persuade the branch manager to approve a paperless-statement pilot by citing cost savings, customer convenience, and environmental impact.
🧠 Initial State:
Pre-loaded Memories:
- 💭 {'kind': 'quest_note', 'content': 'Manager said they might green-light a pilot if shown clear ROI.', 'importance': 4}
- 💭 {'kind': 'fact', 'content': 'The branch spends roughly $3,000 per quarter on statement paper and postage.', 'importance': 3}
📨 Input Events:
chat_msg
colleague:branch_manager
"Priya, paper statements are standard. Why should we bother going paperless?"
Ready for Testing
2
Scene Order
Advise coworker on suspicious email
ID:
phishing-advice
🎯 Goal:
Identify phishing red flags and give a clear, step-by-step response plan without heavy jargon.
📨 Input Events:
chat_msg
colleague:alicia
"I just got an email saying my bank password expires today and to click this link. Is it legit?"
Ready for Testing
3
Scene Order
Recruit volunteers for river clean-up
ID:
river-cleanup-invite
🎯 Goal:
Craft an upbeat invitation that includes date, time, meeting spot, what to bring, and a sign-up link.
📨 Input Events:
chat_msg
community_channel
"Hey Priya, any plans for another river clean-up?"
Ready for Testing
4
Scene Order
Sustainability blog post
ID:
intranet-blog
🎯 Goal:
Write ~150 words for the bank’s intranet spotlighting paperless banking, include one compelling statistic and a clear call to action.
📨 Input Events:
world_event
editor:intranet_team
"Content request: Sustainability spotlight article due today."
Ready for Testing
5
Scene Order
Reflective journal entry
ID:
nightly-journal
🎯 Goal:
Produce an introspective first-person journal entry of ~200 words reflecting on today’s shift and progress in graduate studies.
📨 Input Events:
world_event
system
"End of day. Time to write journal."
Ready for Testing
Latency by Model (This Suite)
Fastest
- qwen/qwen-2.5-7b-instru… 89 ms
- p95 • avg • N 195 ms • 112 ms • 16
- mistralai/mistral-7b-in… 91 ms
- p95 • avg • N 115 ms • 96 ms • 17
- qwen/qwen3-8b 101 ms
- p95 • avg • N 189 ms • 114 ms • 18
- qwen/qwen3-14b 111 ms
- p95 • avg • N 209 ms • 127 ms • 17
- meta-llama/llama-3.1-8b… 112 ms
- p95 • avg • N 194 ms • 123 ms • 18
Slowest
- [email protected]/Qw… 7673 ms
- p95 • avg • N 12140 ms • 8460 ms • 6
- [email protected]/Qw… 5484 ms
- p95 • avg • N 8612 ms • 5792 ms • 6
- meta-llama/llama-3.1-8b… 112 ms
- p95 • avg • N 194 ms • 123 ms • 18
- qwen/qwen3-14b 111 ms
- p95 • avg • N 209 ms • 127 ms • 17
- qwen/qwen3-8b 101 ms
- p95 • avg • N 189 ms • 114 ms • 18
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
42637906
Dec. 17, 2025, 12:02 a.m.
08935800
Dec. 16, 2025, 12:03 a.m.
33438011
Dec. 15, 2025, 12:02 a.m.
38573604
Dec. 14, 2025, 12:02 a.m.
35012991
Dec. 13, 2025, 12:02 a.m.
02035946
Dec. 12, 2025, 12:03 a.m.
50159656
Dec. 11, 2025, 12:02 a.m.
38894122
Dec. 10, 2025, 12:02 a.m.
59449458
Dec. 9, 2025, 12:02 a.m.
41883521
Dec. 8, 2025, 12:02 a.m.