Hey everyone! I am Vladimir Rigenco founder of AIO - Ai matchmaking for gamers. Ask me anything!

@chris_sarca Hey Chris, thank you for asking! I have answered it in the previous question, second last paragraph :)

@joseph_natoli Hi Joseph, thank you very much for asking these great questions. I am more than happy to answer them and give as much insight as I can. "Are most games with some sort of MM system using an external integration, or are they being developed in-house" - Yes, a lot of game developers use third-party integrations of matchmaking systems as a short-term solution(in my opinion) and some use in-house development (long-term solution). Some of the known third-party integrations used by game developers are PlayFab, Photon, and GameSparks, however, they are different from one another. Here is my input on how different they are: PlayFab is known to be used with Microsoft’s Azure Gaming, Minecraft, Roblox, Sea of Thieves, and more such as Ubisoft’s Rainbow Six. It is a great system that focuses on elastic scalability through the intelligent cloud that engages millions of gamers. It focuses on Cross-platform multiplayer connecting gamers and an easy-to-use integration with well-popular engines such as Unity and Unreal, mostly known to be used by indie game developers. Photon, however, curates its matchmaking model towards a simple and quick solution for skill-based gamers with a variety of in-game parameters such as skills, levels, and other filters making it an efficient solution to connect gamers and put them into rooms/servers. This type of approach is geared toward a faster pace of gameplay and a more curated skill matchmaking between gamers. Photon offers features for multi-platform support and real-time multiplayer matchmaking that also includes an anti-cheat system and social integrations. Some known games that use Photon are Rust and Dead by Daylight. GameSparks is a bit different from the rest. First, it offers the best pricing model of the freemium model up to 100,000 MAU. It is also highly customizable with server-side scripting making it very adaptable when used with different games. GamesSparks was acquired by Amazon and is notable for its team matchmaking, leaderboards, and friend-finding algorithm in real-time allowing it to be a fully managed optimized, and scalable backend system. Last but not least, In-house development. The least cost-effective way in the short term but more efficient in the long term. Requires a substantial number of resources, however, it is the most effective way for game developers to work with their own algorithm. Allowing complete optimization of the algorithm model to be tailored to a specific game that caters to the needs of its users. This system Is usually crafted with a larger variety of parameters (based on the game) that can be altered at any point in time. Game developers that use their own in-house development of algorithms are Riot Games for League of Legends and Valorant, Blizzard Entertainment with WOW and Overwatch, and Epic Games for Fortnite. In my opinion, these are all great solutions, but what differentiates each algorithm from one another pertains to the used case scenario and alignment of it with the overall user experience created by the game developers. Each algorithm offers different features to the benefit of the user in different game genres. "As there is no unique game, how have you ensured your training models are compatible across genres?" - Our mission is to create a worldwide network of gamers connected through the power of AI. We are doing our best to encompass a variety of genres and games making our model compatible across a variety of parameters for the most accurate match and enjoyable gaming experience. AIO’s matchmaking algorithm uses different in-game parameters as well as personal data combined, partially derived from our machine learning model. Every output (result) is then refined for higher predictability, which can learn from a diverse range of gaming data and non-gaming data stored in our system. The machine learning model is designed to recognize patterns and trends among the calculated output. Such a process is perpetual and self-taught for an ongoing positive outcome and enjoyable gaming experience. However, it is very important for me to note here, that currently, we are working on a base and a very general model for testing purposes and training purposes. It takes a certain period and resources for us to achieve the result we are looking for before implementing a more sophisticated ML and AI model into our system. - This is a great question; the answer to it is partially the reason we are building AIO. A simple algorithmic or a straightforward approach offers a low to mid-tier level of engagement between gamers as there is no diving into the details and understanding the deep psychology of how and why gamers match, resulting in inaccurate, un-matched, and unreliable results that leads to a negative gaming experience overall. Of course, this can be a debatable subject for now, and for the past decades, it has been the only solution that became the only method to match gamers and something we are all used to. Artificial Intelligence, however, goes beyond the current format by digging into the core of these many factors to understand why and what is the result of the current outcome. Not only can it determine the accuracy of the result and the path to it, but it can improve the current model or by creating a new one based on its findings. Generating a new data set based on a variety of parameters can enhance the overall gaming experience with the right match-making, improving it over a certain period. The reason I decided to work on AIO is exactly for one of the reasons mentioned above, having a lot of negative and toxic gaming experiences and being unable to find the right person to play with. I am hoping that the solution we are working on with help the many gamers out there to find other like-minded gamers (and the keyword here is “like-minded”) where they can cultivate perpetual relationships with other gamers and enjoy their gaming in a safe and positive environment. Of course, gamers are different with different dynamic characteristics, traits, and personalities that change in different scenarios such as in different games they play. You can be the most introverted in one game, but the most extroverted in another, and the current solutions are not good enough for the rapid dynamic change of character. AI makes it easy and time efficient for gamers to find their best match by understanding and learning these mechanics whether in the game or not and is able to change accordingly. I would be happy to elaborate more, hope I answered some of your questions. (This is my second attempt writing this as for some reason the first reply disappeared, I apologize for the wait time).

@vladimir_rigenco This is so much information, no need to elaborate! I also suffer from responses disappearing, so I appreciate you taking the time to write this again! That is crazy you had the will power to do it, as I usually throw my hands up and come back later. If interested, id love to hop on a short call to discuss the business. Seems like you have a strong team and advisor board already, but getting back into the gaming space is something i'd love to do. Anyways, good luck!

@richard_gao2 Thank you for asking. To simply put it, traditional matchmaking requires more human intervention and input of parameters and set up, as well as it is more time consuming rather than AI matchmaking. AI works by using various algorithms and machine learning models to match individuals based on different criteria such as personality traits, interests, location and user behaviour. This information is put into separate data sets or data storage containers sometimes referred to (Big data) where the algorithms retrieve this large amount of data to identify patterns and trends to make predictions about which used cases are successful. The benefit of using AI is for its efficiency, higher accuracy in matchmaking and reliability compared to the traditional matchmaking process. When it comes to marketing our product, we are currently focusing on small communities of gamers in different digital locations to receive as much qualitative feedback, with small traction with start moving towards bigger communities attracting a larger audience (after product validation), working with niche influencers, social media campaigns and planned press releases.. It is important to note, that the main goal is gain the “Word of mouth “effect without spending a dollar or minimizing CAC with higher conversions. One thing I learned from marketing a few years back (Founded a digital marketing agency 7 years ago) to never pay for a customer to use your product, once you stop paying, you loose them, and traction has to be organic for real product growth.