
Technology
|
Gaming Simulation
|
YC W26
|
Valuation:
Undisclosed

Last Updated:
March 24, 2026

An AI-powered grand strategy platform that uses prompt-engineered LLM agents, multi-provider AI routing, and procedural content generation to create dynamic, historically grounded alternate-history simulations with emergent gameplay.
Pax Historia has publicly documented its multi-model AI infrastructure via an Infron AI case study, open-sourced modding APIs and scenario editors on GitHub, and detailed support for 30+ AI providers with unified routing. They've publicly outlined their three-agent architecture (Game Master, Strategic Advisor, Diplomacy Agent) and local inference support via LM Studio and Ollama.
Active GitHub commits point to autonomous play agents using cognitive loops with persistent memory and browser automation, suggesting a push toward RL-based self-play and AI-vs-AI simulation. Architecture choices (stateless agent calls, JSON-driven state, modular design) signal preparation for multiplayer scaling and a potential platform/marketplace model for user-generated scenarios. The multi-provider routing layer via Infron AI indicates sophisticated cost optimization.
Three prompt-engineered LLM agents (Game Master, Strategic Advisor, Diplomacy Agent) operate from a single model instance with dynamically injected historical context to validate actions, generate world events, and simulate diplomacy in real time.
Instead of following a script, three AI characters—a referee, a strategist, and a diplomat—improvise every moment of the game based on real history and what you just did.
It's like having three improv actors—a judge, a war room advisor, and a foreign ambassador—who've memorized every history textbook and never repeat the same scene twice.
A unified API layer routes LLM inference across 30+ providers with intelligent model selection based on cost, latency, and throughput, plus automatic failover during outages.
A smart traffic controller automatically picks the cheapest, fastest AI brain available for each task and instantly switches to a backup if one goes down.
It's like a travel booking engine that checks every airline in real time, picks the cheapest flight that still lands on time, and automatically rebooks you if your plane gets cancelled—except for AI brainpower instead of seats.
Experimental autonomous agents use a cognitive loop—perceiving game state, reasoning with persistent memory, and executing actions via browser automation—to play the game independently using LLM-driven strategic planning.
A robot player that can see the game board, remember its long-term plans, and actually click buttons to play the entire game by itself—learning strategy like a human would.
It's like building a chess engine that doesn't just calculate moves but actually sits at your computer, reads the board with its eyes, remembers why it sacrificed that pawn six turns ago, and clicks the mouse to play.
Pax Historia's multi-provider AI routing layer combined with prompt-engineered agent specialization allows it to run the same game across 30+ LLM providers with automatic failover, cost optimization, and model switching, something no competitor in the strategy gaming space currently offers.