> For the complete documentation index, see [llms.txt](https://docs.altura.com/altura-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.altura.com/altura-documentation/altura-intelligence/npc-engine/npc-engine-+.md).

# NPC Engine +

## Overview

\
NPC Engine+ builds upon the NPC Engine by adding a persistent, player-specific memory layer. This creates individually tailored character arcs, making each player’s experience unique and reactive to their personal journey.

## Model Type & Architecture

* Enhanced Memory and Context Layer: Alongside the general memory store, NPC Engine+ maintains a dedicated memory embedding store keyed per player. This allows NPCs to remember specific player actions, choices, and dialogue history, enabling personalized storylines and unique character relationships.
* Long-Term Adaptation: Reinforcement learning techniques are extended to track and adapt to each player’s evolving playstyle. NPC Engine+ continuously refines character responses over multiple sessions, incorporating the player’s past decisions to drive narrative branches.

## Underlying Tech

* Personalized Vector Databases: Each player’s interactions are stored as embeddings, enabling highly personalized retrieval-augmented generation.
* RL with User Modeling: RL agents learn policies that maximize player satisfaction metrics, narrative depth, and player retention over long story arcs.
* Multi-Session Continuity: Persistent storage mechanisms ensure that even if a player returns after weeks, the NPC remembers their last interaction and can build upon it.