At a Glance
- Tasks: Train superhuman game agents and transform their intelligence into general models.
- Company: Join a cutting-edge research team focused on AI and gaming.
- Benefits: Remote-first work, competitive salary, and opportunities for publication.
- Other info: Collaborative environment with regular in-person meetups across Canada and the US.
- Why this job: Make a real impact in AI by proving games enhance model capabilities.
- Qualifications: Experience building game agents and hands-on RL knowledge required.
The predicted salary is between 135000 - 165000 $ per year.
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 Alpha Star, Alpha Go, or Open AI 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.
- #J-18808-Ljbffr
Research Scientist in Ottawa employer: Good Start Labs
Join a forward-thinking team that values innovation and collaboration, where your work as a Research Scientist will directly contribute to groundbreaking advancements in AI through gaming. Enjoy a remote-first work culture that fosters flexibility and creativity, complemented by regular in-person gatherings to strengthen team bonds. With ample opportunities for professional growth and a commitment to publishing impactful research, this is an ideal environment for those passionate about pushing the boundaries of technology.
