game testing

Filuta AI

Ensure bug-free releases with AI-powered planning agents that autonomously explore games, execute complex goals, and adapt to updates to slash QA costs by 70%.

Filuta AI screenshot

About Filuta AI

Filuta AI is an advanced automated testing framework designed to streamline quality assurance processes within the video game industry. Utilizing a specialized form of "Composite AI," the platform employs autonomous planning agents that function as virtual players. These agents are built on logic and technology originally developed for complex autonomous missions, such as deep-space exploration and deep-sea navigation. Unlike traditional automation tools that rely on fragile, hard-coded scripts, Filuta AI utilizes a goal-oriented approach. Developers define high-level objectives—such as reaching a specific location, defeating an enemy, or interacting with environment objects—and the AI autonomously determines and executes the necessary sequence of actions to achieve those results while navigating the game’s mechanics in real-time. The core of the system relies on "Digital Twin" instrumentation, where the game provides the agents with the necessary data to understand its mechanics, available actions, and environment states. This allows for adaptive exploration, where the AI agents can uncover hidden bugs, edge cases, and unintended behaviors that manual testers might overlook. One of the most significant advantages of this approach is that the tests are self-maintaining. As the game’s world or mechanics change during development, the agents re-calculate their plans based on the new state rather than breaking. This eliminates the need for QA engineers to constantly rewrite test scripts, allowing the system to provide consistent 24/7 testing coverage. This solution is primarily designed for game studios of all sizes, ranging from independent developers to large-scale enterprise teams. It specifically addresses the needs of QA Leads, Game Designers, and Automation Engineers who require a more efficient way to handle repetitive testing tasks. By automating the "heavy lifting" of playtesting, Filuta AI enables these professionals to focus on high-level strategy and creative goal-setting rather than manual grinding. The platform is entirely engine-agnostic, offering native support for Unity and Unreal Engine, as well as the flexibility to integrate with custom internal engines used by large studios. What distinguishes Filuta AI from other automated testing tools is its deep roots in academic research and specialized engineering. The team consists of over 30 members, including 16 PhDs and three professors with extensive backgrounds in mathematics and artificial intelligence research. This expertise translates into a tool that produces detailed, actionable bug reports complete with reproduction steps and performance metrics. By claiming to reduce overall QA costs by up to 70% and significantly shortening development cycles, Filuta AI offers a sophisticated alternative to manual testing labor, providing a scalable way to ensure game stability from early prototypes through to the final release.

Filuta AI pros & cons

Pros

  • Reduces QA costs by up to 70% compared to traditional manual testing.
  • Eliminates the need for manual test script maintenance as the game evolves.
  • Provides continuous 24/7 testing coverage without human intervention.
  • Supports Unity, Unreal, and custom engines for maximum flexibility.
  • Developed by a world-class team including 16 PhDs in AI and mathematics.

Cons

  • Requires initial game instrumentation to provide agents with a Digital Twin.
  • Public pricing details are not available without scheduling a demo.
  • May require a technical pilot program to establish a proof of concept.

Filuta AI use cases

  • QA Leads can implement 24/7 automated testing to uncover edge-case bugs that are often missed during manual testing cycles.
  • Game Designers can set specific objective goals to verify that puzzles and quest logic are solvable across all intended paths.
  • Automation Engineers can reduce their workload by using adaptive agents that don't require constant script updates when game maps change.
  • Producers can shorten development cycles by receiving consistent, detailed performance metrics and bug reports in real-time.

Filuta AI features

  • adaptive exploration
  • goal-driven testing
  • detailed bug reproduction steps
  • 24/7 testing environment
  • engine-agnostic support
  • composite ai technology
  • digital twin instrumentation
  • autonomous planning agents

Filuta AI FAQs

What game engines does Filuta AI support?

Filuta AI is completely engine-agnostic, meaning it can be integrated into any architecture. It provides native support for popular engines like Unity and Unreal, as well as any custom internal engines.

How does Filuta AI handle game updates and mechanical changes?

The AI agents use an adaptive framework that analyzes the game state in real-time to update their plans. This eliminates the need for manual test script maintenance when the game world or mechanics evolve.

What kind of test goals can I define for the AI agents?

Users can set specific goals such as 'reach the castle from the starting village' or 'fail the quest'. The agents then autonomously plan and execute the necessary actions to achieve those specific outcomes.

What information is included in the bug reports?

Filuta AI provides detailed actionable insights, including specific reproduction steps and clear performance metrics. This allows development teams to quickly identify and resolve issues found during autonomous play.

Open roles

All AI jobs

Neurosymbolic Systems Engineer

Education Requirements:

  • Master's or PhD degree in math, computer science, statistics or a related field

Experience Requirements:

  • Proficiency in PyTorch/TensorFlow and symbolic AI tools

  • Hands-on experience with transformer architectures and convolutional networks

  • Experience in MLOps (building and deploying machine learning models in production)

  • Experience working with data ingestion and cleaning tools

  • Expertise in model evaluation

Other Requirements:

  • Fluency in Python, with experience deploying hybrid AI systems via Docker/Kubernetes

  • Strong foundation in classic AI/ML: Bayesian networks, graph algorithms, and optimization techniques

  • Familiarity with neuro-symbolic integration patterns

  • Experience with automated reasoning frameworks

  • Experience with distributed computing environments and tools

  • Experience with automated system deployment and orchestration for learning

  • Publications or projects demonstrating hybrid AI systems

Responsibilities:

  • Contribute to the design of neurosymbolic architectures

  • Implement RAG pipelines using vector/vector-infused databases

  • Optimize transformer models via LoRA and deep fine-tuning

  • Develop novel methods to convert neural network outputs into symbolic representations

  • Evaluate system performance using metrics for accuracy, robustness, and explainability

Show more details

Ratings & reviews

No reviews yet. Be the first to share how Filuta AI worked for you.