Senior ML Engineer - Real-Time Unity AI Agents (Equity)

ironSource; Montreal (administrative region); On-site; CAD 136,000 - 221,000; Full time

Benefits offered by this job

Equity grants

Annual discretionary bonus

Job summary

ironSource is seeking a senior ML/AI engineer to design, implement, and scale AI agent systems in Unity, enabling goal-directed behavior, task planning, and environment-aware decision making for real‑time gameplay experiences.

You will develop reusable agentic frameworks and integrate ML models within strict latency constraints, collaborating across teams to push Unity's agentic AI offerings and mentor other engineers and researchers.

Qualifications

• Advanced degree in CS/ML/Robotics or equivalent practical experience.
• 5+ years of hands-on experience developing ML systems for real-time environments.
• Strong background in reinforcement learning, AI planning, or multi-agent systems.

Responsibilities

• Design, implement, and scale AI agent systems in Unity for goal-directed behavior.
• Develop reusable agentic frameworks and memory/policy modules for creators.
• Integrate ML models within real-time constraints to optimize latency and determinism.
• Lead high-impact initiatives in hierarchical RL, planning, and world modeling.
• Collaborate across teams to unlock innovative use cases in gaming and simulations.
• Contribute to the long-term vision and roadmap for Unity's agentic AI offerings.
• Establish best practices for model quality, evaluation, and AI safety.
• Mentor engineers and researchers to foster technical excellence.

Skills

Reinforcement learning

Decision-making under uncertainty

AI planning

Multi-agent systems

Education

MS/PhD in Computer Science, ML, Robotics or related field

Job description

ironSource is seeking a senior ML/AI engineer to design, implement, and scale AI agent systems in Unity, enabling goal-directed behavior, task planning, and environment-aware decision making for real‑time gameplay experiences.

You will develop reusable agentic frameworks and integrate ML models within strict latency constraints, collaborating across teams to push Unity's agentic AI offerings and mentor other engineers and researchers.