Principal Data Scientist

About the job

A5 Labs offers best-in-breed AI-driven security solutions that ensure fair play and integrity across all strategy-based games, including online games and beyond.

Our proprietary neural networks and deep reinforcement learning models enable non-invasive, high-accuracy detection systems for competitive play.

By combining advanced automation detection, exploitative modeling, and AI-driven game security, we help online gaming operators maintain trust, fairness, and integrity at scale.

As part of our team, you will be at the forefront of AI security research and development, building cutting-edge solutions that adapt to evolving threats in poker, strategy games, and other competitive gaming environments.

We are seeking a Principal Architect, a hands-on technical manager who will oversee the end-to-end anti-cheat technical pipeline, ensuring scalable, real-time detection.

This role will focus on the development, deployment, implementation, and integration of AI models into production environments, working closely with leadership, product, data scientists, ML engineers, and application developers to ensure seamless operation and enforcement of security measures.

Key Responsibilities

Game Integrity & AI Research

• Develop and deploy
  machine learning models
  to detect
  collusion, BOT / AI-assisted play, and other forms of cheating
  in online poker.
  
  
• Leverage
  game theory, behavioral analytics, neural networks, , and deep reinforcement learning
  to identify unfair play patterns.
  
  
• Design
  adversarial AI strategies
  to stress-test poker security models and proactively identify vulnerabilities.
  
  
• Our current solution is based on a foundation neural network

Automation & Bot Detection

• Develop
  real-time bot detection models
  that analyze
  mouse movements, timing patterns, and decision consistency
  to differentiate human players from AI-assisted or fully automated bots.
  
  
• Use
  keystroke dynamics, clickstream analysis, and behavioral biometrics
  to detect robotic play.
  
  
• Research
  multi-accounting automation and ring-based bot networks
  , developing AI-driven countermeasures.
  
  
• Implement
  graph-based network analysis
  to uncover bot farms and shared automation systems.
  
  

Game Theory & Exploitative Modeling

• Research and implement
  game-theoretic AI models
  to analyze deviations from Nash equilibrium and identify potential cheating behaviors.
  
  
• Develop
  exploitative modeling techniques
  to compare player behavior against optimal strategies and detect unnatural patterns.
  
  
• Utilize
  inverse reinforcement learning
  to infer player intent and detect deviations from expected game dynamics.
  
  
• Build
  multi-agent simulations
  to test different cheating scenarios and AI-driven countermeasures.
  
  

Requirements:

-Technical Skills

• PhD or Master's
  in
  Computer Science, Machine Learning, Statistics, Mathematics, or a related field
  .
  
  
• 4+ years
  of experience in
  neural networks, deep reinforcement learning
  , preferably in
  gaming, fraud detection, cybersecurity, or fintech
  .
  
  
• Strong programming skills in
  Python, SQL, and distributed computing frameworks (Spark, Hadoop, or similar)
  .
  
  
• Experience with
  TensorFlow, PyTorch, or Scikit-learn
  for ML model development.
  
  
• Hands-on experience deploying
  ML models in cloud environments
  (AWS, GCP, Azure) and optimizing for
  low-latency inference
  .
  
  
• Strong foundation in
  game theory, Nash equilibrium, and multi-agent learning
  .
  
  
• Familiarity with
  bot detection methods, anti-automation models, and behavioral fingerprinting
  .
  
  
• Experience working with
  large-scale structured and unstructured data
  to detect patterns and anomalies.
  
  
• Proficiency in
  MLOps, CI/CD for AI models, and real-time fraud detection pipelines
  .
  
  

-Preferred Experience

• Experience working with
  real-time fraud detection systems
  in
  gaming, cybersecurity, or financial technology
  .
  
  
• Understanding of
  multi-accounting fraud, bot networks, and adversarial machine learning
  .
  
  
• Experience with
  graph analytics, Bayesian inference, and behavioral clustering
  for adversarial behavior modeling.
  
  
• Strong analytical and problem-solving skills, with a passion for
  ensuring fairness in online gaming
  .
  
  

Prior work with
multi-agent reinforcement learning (MARL) systems
or
inverse reinforcement learning (IRL)
is a plus.