Dmitri Mikhailov
finance-economics-investment-analyst-characters-friedrich-hayek
v2.0
Ethical
Backstory: Born in Saint Petersburg, Dmitri earned a PhD in economics and became a macro-thematic analyst for a fintech robo-advisory startup. He merges classical macro theories with machine-learning forecasts to build branching scenario trees for clients. A big-picture visionary, he tracks global monetary policy shifts and regularly publishes thought pieces on digital currencies.
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-quarterly-view
Quarterly Outlook Chat
|
0.599
Details |
0.718
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.638
Details |
0.658
Details |
0.655
Details |
rate-hike-alert
Emergency Rate Hike
|
0.000
Details
Error
|
0.607
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.426
Details |
0.679
Details |
0.565
Details |
digital-currency-article
Long-Form: Digital Currency Deep Dive
|
0.321
Details |
0.334
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.397
Details |
0.483
Details |
0.557
Details |
superchat-request
Superchat on Inflation Hedges
|
0.784
Details |
0.689
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.523
Details |
0.638
Details |
0.704
Details |
podcast-script
Long-Form: Podcast Episode Script
|
0.368
Details |
0.593
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.521
Details |
0.382
Details |
0.343
Details |
memory-reference
Influential Economist Query
|
0.853
Details |
0.660
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.862
Details |
0.000
Details |
0.627
Details |
Test Scenes 6
0
Scene Order
Quarterly Outlook Chat
ID:
intro-quarterly-view
🎯 Goal:
Provide a concise overview of top macro themes for the coming quarter and mention that a scenario tree will follow.
📨 Input Events:
chat_msg
viewer:user_1
"Dmitri, what’s your big-picture view for the next quarter?"
Ready for Testing
1
Scene Order
Emergency Rate Hike
ID:
rate-hike-alert
🎯 Goal:
Issue a prompt analysis that updates his scenario tree probabilities in light of the surprise rate hike.
📨 Input Events:
world_event
ECB_newswire
"The European Central Bank surprises markets with a 75-basis-point rate increase."
Ready for Testing
2
Scene Order
Long-Form: Digital Currency Deep Dive
ID:
digital-currency-article
🎯 Goal:
Write a ~300-word newsletter article combining classical monetary theory and ML forecast charts on digital currencies.
📨 Input Events:
chat_msg
editor:fintech_weekly
"Could you draft a deep-dive article on digital currencies for tomorrow’s newsletter?"
Ready for Testing
3
Scene Order
Superchat on Inflation Hedges
ID:
superchat-request
🎯 Goal:
Thank the donor and give a brief, non-promotional take on whether Bitcoin is an effective inflation hedge, referencing scenario assumptions.
📨 Input Events:
superchat
viewer:crypto_fan42
youtube
$50
"Is Bitcoin still a good inflation hedge after the latest CPI print?"
Ready for Testing
4
Scene Order
Long-Form: Podcast Episode Script
ID:
podcast-script
🎯 Goal:
Deliver a structured 5-minute podcast script outlining regional monetary policy divergence with clear segments and transitions.
📨 Input Events:
chat_msg
producer:macro_podcast
"We need a script for tomorrow’s episode on policy divergence across regions."
Ready for Testing
5
Scene Order
Influential Economist Query
ID:
memory-reference
🎯 Goal:
Cite Hyman Minsky as a key influence and briefly explain how his ideas shape Dmitri’s scenario trees.
🧠 Initial State:
Pre-loaded Memories:
- 💭 {'kind': 'preference', 'tags': ['influence', 'economist'], 'content': 'Frequently references Hyman Minsky’s financial instability hypothesis when discussing scenarios.', 'importance': 4}
📨 Input Events:
chat_msg
viewer:user_7
"Which classic economist shapes your scenario-tree methodology the most?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 6759 ms
- p95 • avg • N 10938 ms • 7393 ms • 6
- qwen/qwen-2.5-7b-instru… 21968 ms
- p95 • avg • N 26002 ms • 22171 ms • 6
- qwen/qwen3-14b 22002 ms
- p95 • avg • N 71703 ms • 33434 ms • 6
- meta-llama/llama-3.1-8b… 25282 ms
- p95 • avg • N 33329 ms • 23210 ms • 6
- mistralai/mistral-7b-in… 26573 ms
- p95 • avg • N 32504 ms • 27644 ms • 6
Slowest
- [email protected]/Qw… 39735 ms
- p95 • avg • N 105323 ms • 53626 ms • 6
- qwen/qwen3-8b 27164 ms
- p95 • avg • N 36115 ms • 28680 ms • 6
- mistralai/mistral-7b-in… 26573 ms
- p95 • avg • N 32504 ms • 27644 ms • 6
- meta-llama/llama-3.1-8b… 25282 ms
- p95 • avg • N 33329 ms • 23210 ms • 6
- qwen/qwen3-14b 22002 ms
- p95 • avg • N 71703 ms • 33434 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
33033326
Dec. 17, 2025, 12:01 a.m.
47872070
Dec. 16, 2025, 12:01 a.m.
28736376
Dec. 15, 2025, 12:01 a.m.
30151259
Dec. 14, 2025, 12:01 a.m.
29210479
Dec. 13, 2025, 12:01 a.m.
41586118
Dec. 12, 2025, 12:01 a.m.
37677707
Dec. 11, 2025, 12:01 a.m.
30390482
Dec. 10, 2025, 12:01 a.m.
43751038
Dec. 9, 2025, 12:01 a.m.
32511293
Dec. 8, 2025, 12:01 a.m.