> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nebulaengage.com/llms.txt
> Use this file to discover all available pages before exploring further.
# How Nebula Simulations Work
> A Nebula simulation is an AI-driven session inside a fictional company. Learn how sessions start, progress, and end — and what persists between runs.
A Nebula simulation is more than a conversation — it's a structured engagement inside a living corporate world. Each time you start a session, you step into a specific role at one of Nebula's 12 fictional companies, receive a case study tailored to your experience level, and work through real workplace pressures: competing priorities, demanding stakeholders, and decisions that actually matter. The world remembers who you are, but every scenario is generated fresh.
## What a simulation session is
A session is a single, timed engagement inside a persistent company world. It has a clear beginning (onboarding and role assignment), a middle (active work in the virtual desktop), and an end (a scored session report). Sessions are anchored to a specific company, domain, case study, and difficulty level — all set during onboarding.
Sessions are not generic chat experiences. They run inside a structured simulation engine that tracks world state, persona relationships, your decisions, and your progression in real time.
## Session lifecycle
Every session passes through a defined sequence of statuses.
The session is being created. Nebula loads your selected company's world state, attaches your user context, and prepares the onboarding coordinator.
You chat with an AI coordinator who assesses your experience level, identifies whether you're a student or a professional, and assigns you a role and case study. An optional baseline assessment is offered here.
You're inside the virtual desktop simulation workspace. You interact with personas, read and write emails, complete tasks, manage documents, and navigate decision checkpoints as the scenario evolves.
The session is temporarily suspended. Your progress is preserved exactly where you left off.
The session has ended normally. A full session report is generated, including scores, skill updates, XP gains, and replay data.
The session ended without completion. Partial progress may still contribute to your progression depending on what was logged before exit.
## Key session components
A dynamically named AI guide who runs you through the onboarding flow at the start of every session. They infer your background, suggest a difficulty level, propose three case studies, and assign your role.
The simulation environment where your session plays out. It includes email, team chat, a document editor, task management, and access to Lucy AI — your in-simulation AI advisor.
AI-powered coworkers, managers, and stakeholders you interact with throughout the session. Each persona has a distinct personality, communication style, and relationship history that persists across sessions.
Structured moments in the scenario where you're asked to make a meaningful choice. Your responses are evaluated and scored, feeding directly into your skill progression.
A real-time briefing hub inside the workspace that tracks your objectives, stakeholder map, constraints, milestones, and decision log throughout the session.
A detailed performance summary generated at session end. It includes your overall score, skill scores, XP earned, readiness signal, decision patterns, and personalized insights.
## Difficulty levels
When you start a simulation, the onboarding coordinator suggests a difficulty level based on what it infers about your background. You can confirm or adjust the suggestion.
| Level | Who it's for |
| -------------- | ------------------------------------------------------- |
| `beginner` | Early students or newcomers to a domain |
| `intermediate` | Students or professionals with some relevant experience |
| `advanced` | Experienced professionals looking for a real challenge |
| `expert` | Senior professionals seeking executive-level complexity |
Difficulty affects the complexity of the case study, the demands of personas, the pace of chaos events, and the scoring bar for decision checkpoints.
## What persists vs. what resets
Nebula separates its world into a **stable layer** that persists across all sessions and a **dynamic layer** that changes over time.
**Persists across sessions:**
* Company identity, culture, org structure, and reporting lines
* Persona memory — coworkers remember your prior interactions
* Your skills, XP, credits, and progression profile
**Resets or evolves each session:**
* The active case study and scenario state
* Market conditions, internal politics, and business priorities
* Decision checkpoint outcomes
* Session-specific objectives and deliverables
This means returning to the same company feels familiar — the same people, the same culture — but the challenge you face is always fresh.
## Dynamic chaos and disruption events
During an active session, Nebula injects realistic workplace friction. These **disruption events** are triggered by time, your decisions, or scenario conditions — not by a fixed script.
Examples of what can happen mid-session:
* A key deadline moves up unexpectedly
* A stakeholder reverses a prior commitment
* Budget constraints are suddenly tightened
* A competing priority pulls resources from your project
* A crisis lands that requires an immediate response
Disruptions are designed to test your judgment under pressure, not just your ability to follow a plan.
## Event logging and session replay
Every meaningful action in a session — decisions made, personas responded to, chaos events triggered, objectives completed — is logged as a structured event. This event log powers two important features:
* **Replay:** After a session ends, you can review key moments and understand how the simulation unfolded.
* **Branching:** At certain checkpoints, you can explore alternative outcomes and see how a different decision would have played out.
The event log is also the source of truth for your session report and skill scoring.
Replay data is preserved after a session ends, so you can return to review your performance or explore alternate paths at any time.