AI Dynamic Difficulty Adjustment Engine Developer

Pelvex Tech

Remote · $160k – $200k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

The ideal difficulty curve in a game keeps the player in flow state: challenged enough to stay engaged, not so overwhelmed that they quit in frustration, and not so under-challenged that they disengage out of boredom. Static difficulty settings set at the start of a session can't maintain that balance because player skill varies not just between players but within the same session as a player warms up, gets tired, or encounters a new mechanic. Dynamic Difficulty Adjustment adapts game parameters in real time based on inferred player state — tracking frustration signals (repeated failures, long hesitation times, error patterns) and boredom signals (rapid completion, minimal exploration, declining engagement) and adjusting enemy AI difficulty, puzzle complexity, or resource availability to bring the player back toward flow. We build game AI systems for 12 live service games. DDA is the most requested capability from game design teams: current implementations are rule-based and crude. You'll build the ML-driven DDA engine: designing the player state inference model from behavioral telemetry, building the difficulty parameter control system for different game genres, and running the reinforcement learning-based policy that learns which adjustments most reliably return players to engagement.

Fully remote.

Responsibilities

Design player state inference model from behavioral telemetry: frustration, boredom, and flow state estimation
Build difficulty parameter control system adapting multiple game variables across different genre mechanics
Implement RL-based adjustment policy learning from player engagement signals which adjustments are most effective
Develop session-level difficulty curve management that maintains challenge arc across multi-hour sessions
Create DDA evaluation framework measuring player retention and session length as primary adjustment quality metrics
Build per-player adaptation that learns individual difficulty preferences from session history

Requirements

6+ years in game AI or ML engineering with player modeling, behavioral analytics, or RL experience
Deep understanding of game design principles, difficulty design, and player engagement psychology
Strong Python and PyTorch skills with reinforcement learning and behavioral time-series modeling
Experience deploying adaptive AI systems in live service games with real-time requirements
Track record demonstrating player retention or engagement improvements from AI difficulty systems