ManaMind, an AI company building autonomous testing agents for game development, has raised $1.5 million in pre-seed funding from Sure Valley Ventures, EWOR, Ascension, SyndicateRoom, and HEARTFELT. The company was founded by Emil Kostadinov and is developing AI agents that autonomously play games to detect bugs that human testers miss, targeting one of the most persistent and expensive pain points in the game development lifecycle.
The core of the platform is HiveMind, ManaMind’s proprietary AI model, which the founders say has outperformed models from Google, OpenAI, and Anthropic on internal benchmarks. Unlike traditional quality assurance processes that rely on human testers following scripted test plans, ManaMind’s agents explore game environments dynamically, simulating real player behavior and uncovering edge cases and failure states that scripted or manual testing routinely overlooks. The result is a testing process that is faster, more comprehensive, and able to operate continuously without the fatigue, inconsistency, or cost constraints that limit human QA teams.
The timing reflects a broader inflection point in game development. Modern games ship with enormous complexity, sprawling open worlds, procedurally generated content, and intricate multiplayer systems that generate near-infinite combinations of player states. The scale of that complexity has outpaced the ability of traditional QA pipelines to keep up, contributing to a wave of high-profile launches marred by bugs and instability. ManaMind positions its platform as the infrastructure layer that can finally bring testing capacity in line with development ambition.
Gaming is ManaMind’s entry point, but the company has articulated a longer-term ambition that extends well beyond it. The founders view autonomous testing as a fundamental need across all software development, and plan to use the gaming market as the proving ground for a platform they intend to scale across the broader software industry. Games, with their highly interactive environments, unpredictable user behavior, and real-time rendering demands, represent one of the most challenging testing environments in software development. Successfully automating QA at that level of complexity would validate the platform’s capabilities for virtually any software application downstream.
The pre-seed round will support further development of the HiveMind model, expansion of the platform’s capabilities, and early customer acquisition within the gaming industry.
KEY QUOTE:
“Manual QA is broken. It’s expensive, slow, and can’t keep up with modern game development. Our AI agents play through games autonomously and catch bugs human testers miss. Gaming is the launchpad. Autonomous testing for all software is the mission.”
Emil Kostadinov, Founder and CEO, ManaMind
