Create Your Own Synthetic Employee
Capture how a specific person thinks, communicates, and makes decisions — and put it on demand.
Institutional knowledge walks out the door when people leave. Onboarding takes time. And some expertise — a founding engineer's system intuition, a top seller's objection-handling instincts — is nearly impossible to document.
iMario lets you build a synthetic individual that captures how a specific person thinks, communicates, and makes decisions.
When to Use This
- You want to preserve the knowledge and judgment of a key team member
- You're onboarding new hires and want them to have someone to learn from on demand
- You want to simulate how a specific colleague or expert would respond to a situation
Step 1: Build Your Individual
Use Solo Mode.
You have two options:
From LinkedIn Paste a LinkedIn profile URL to generate an individual grounded in their professional background, experience, and communication style.
From a Personality Test Have the person talk to the AI analyst for approximately 8 minutes. Their responses are used to build a synthetic individual that reflects their thinking style, personality, and reasoning patterns.
For the most accurate result, use both — LinkedIn for professional context, personality test for behavioral depth.
Step 2: Configure Your Audience Node
In the Audience Node, select the individual you just created.
This Playbook is built around a single Individual in Solo Mode. There's no segment comparison or multi-flow setup — the focus is on depth and accuracy for one specific person.
Step 3: Extend with Knowledge Bases and Agent Skills
A synthetic employee becomes more useful the more context you give it.
Connect a Knowledge Base to ground them in specific domain knowledge:
- Internal documentation
- Past decisions and rationale
- Product specs or process guides
Connect Agent Skills to extend what they can do:
- Apply specialized reasoning
- Execute structured tasks
- Interact with defined workflows
The combination of personality, memory, and knowledge makes the individual genuinely useful — not just a chatbot with a name.
Step 4: Interact via Chat
There's no Task Canvas for this Playbook. Use Chat to interact directly.
Ask them what you'd ask the real person:
- "How would you approach this architecture decision?"
- "What objections should I expect from this type of buyer?"
- "How did we handle this kind of situation before?"
Each interaction updates their memory, making them more accurate and useful over time.
What to Look For
A well-built synthetic employee should:
- Respond in a recognizable voice — if people who know the real person find the responses plausible, it's working
- Surface non-obvious knowledge — not just facts, but judgment calls and reasoning patterns
- Improve with use — memory means every conversation makes the individual more grounded
This isn't a replacement for the real person. It's a way to make their knowledge available at scale.
See also
- Set Up Your Task — configure Content, Audience, and Output for this playbook
- Read Your Report — interpret the qualitative findings
- How Many Synthetic Individuals Do I Need? — sizing guidance for the audience