RESEARCH SCIENTIST

Good Start Labs; Ottawa; Hybrid; CAD 90,000 - 130,000; Full time

Job summary

Good Start Labs is recruiting for a research-focused role in Ottawa that combines game AI with scalable language-model distillation. You will train superhuman game agents, distill insights into general models, and design benchmarks using the game engine as a verifier.

The work emphasizes RL post-training methods (PPO/GRPO/RLVR), designing rewards for multi-agent interactions, and producing publishable results that advance the lab’s research brand.

Qualifications

• Experience building or distilling game agents into general models.
• Public research record with papers, models, or benchmarks.
• Ability to design experiments and derive conclusions from messy results.

Responsibilities

• Train expert game agents and distill their intelligence into general language models.
• Design dense reward signals and shape what gets learned.
• Build training systems that mirror a benchmark and use the game engine as verifier.
• Run supervised fine-tuning and RL experiments measuring game environments impact on model performance.
• Design rewards/evaluations for multi-agent games including negotiation and cooperation.
• Publish papers, technical reports, and blog posts to bolster research brand.

Skills

Reinforcement learning

PPO/GRPO/RLVR

Game AI

Research experience

Publications/ benchmarks

Tools

Python

PyTorch

TensorFlow

Job description

The job: prove that games make models better.

You build expert agents that play games at a superhuman level, then distill that intelligence into general models. You isolate what game data and game-based RL do to model capability, and you make the transfer repeatable.

What You'll Do

• Train expert game agents and distill their intelligence into general language models.
• Design dense reward signals and shape what gets learned. Our first transfer result came from sparse rewards and expensive rollouts. You help improve that process.
• Build training systems that target a benchmark: generate tasks that mirror its format and use the game engine as the verifier.
• Run controlled SFT (supervised fine-tuning) and RL experiments that measure how game environments change model performance.
• Design rewards and evaluations for multi-agent games: negotiation, long-horizon planning, cooperation, deception.
• Build public evaluations and benchmarks that show what games measure and math or code benchmarks miss.
• Publish papers, technical reports, and blog posts. Your results will help carry our research brand.
• Feed findings back into environment design with the engineering team.

What We Look For

• The profile we want most: you built a superhuman game agent, in the spirit of AlphaStar, AlphaGo, or OpenAI Five. Any company, any game, show us the agent and what it beat.
• Down for the mission. You love games and you view them as serious training grounds for intelligence.
• Hands-on RL post-training experience: PPO, GRPO, or RLVR.
• A public research record: papers, models, or benchmarks that other people used, cited, or built on.
• Designs small, fast experiments and pulls real conclusions from messy results.

Comp and Location

• Remote-first with regular in-person gatherings. Must be in the US or Canada.