Session: AI Effortless Game Testing With Planning Agents
summary
Filuta AI presents a planning-agent QA system that generates goal-driven test sequences instead of scripted steps, claiming 17.5x testing value per dollar and citing per-line test-code costs of about $10 to create and $3/yr to maintain. Agents model whole game systems (quests, dialogue, monetization), run cloud-scaled combinatorial solvers to find regressions and gameplay anomalies, produce logs and video, and integrate with engines via a Unity plugin or Unreal/custom engine interfaces; on-prem deployment exists for defense customers but cloud is the default.
AI-written summary · about
chapters (12)
- 0:00AI game testing intro
- 1:27Modern QA testing challenges
- 3:14Planning agents for testing
- 5:24Agent benefits and insights
- 6:29Company background
- 7:56Symbolic vs sub-symbolic AI
- 9:22Symbolic AI advantages
- 10:03How planning agents work
- 11:25Engine integration and testing types
- 13:13Games for autonomous testing
- 13:57Client testimonial impact
- 15:02Q&A
related talks
21mEffortless Game Testing With Planning Agents
22mBeyond Bugs: How AI turns QA into Full-Circle Engineers
43mTesting Agentic AI Applications: Beyond Traditional QA
37mAgentic AI and the Next Decade of Quality Engineering
42mIt's Inevitable, Mr. Anderson: Machines, Humans, and the Future of Testing
29mQA vs AI: Friends or Foes?
30mAgentic QA vs LLM Dev: A Live AI Showdown in the DevOps Arena
32mQuality engineering in the AI age: From testing AI to leveraging GenAI