Image


About

Welcome! I'm Aleksei (Lex) Andreev, a software engineer based in Singapore (PR). This website serves as a personal portfolio showcasing my work and technical expertise. Nice to meet you!
Shader Alert: If you notice any elements flickering, lagging, or appearing incorrectly, try viewing the website in a different browser.
I'm a Senior Software Engineer with over a decade of experience in the game industry.
My skillset spans a broad spectrum, shaped by both the diverse demands of the industry and my own explorative approach to problem-solving.
I have worked on developing production tools and pipelines, implementing various aspects of game logic, designing game AI systems, rendering systems, VFX systems, post-processing, procedural generators, audio controllers, GUI systems from scratch, as well as building system infrastructures, data flows, and other architectural components. I have also been involved in management, hiring, and team-building efforts, though my strengths lie more in technical problem-solving. Additionally, I've implemented numerous API adaptations tailored to diverse requirements.
I have a particular passion for areas I’ve delved into more deeply, including rendering challenges, mathematical optimization, general-purpose GPU programming (such as compute shaders), simulations, and procedural generation. I am also interested in machine learning research, having experimented with genetic algorithms, Markov chains, and related techniques.

A significant portion of my career was spent at Kwest LLC, where I contributed extensively to various projects. The primary product was a location-based AR storytelling game called "Kwest Go".

Kwest Go Demo (Fullscreen Recommended)

To highlight my contribution, the CEO of Kwest, Omeed Shams, remarked that the game would not be what it is without my work—if it would exist at all.

One of my key contributions was "Kwest Graph"—a production toolset that enabled the creation of the entire experience without the need for programmers (though, like any tool, it required continuous refinement and adjustments).

Another achievement I'm proud of is the map rendering system. We partnered with Google and utilized their "MUSK" API, which posed challenges due to the massive number of meshes and high polycount (hundreds of thousands) that led to thousands of batches. To address these issues, I developed an alternative rendering system. Here are the details. Looking back, there are areas I would approach differently for greater efficiency.

Unfortunately, the project came to a halt when the Covid-19 outbreak made location-based games less viable, leading us to pivot to other projects. I then joined "Room 40 Games," where we focused on quickly developed hyper-casual mobile games. One of the games I worked on stands out as particularly enjoyable to develop, and I'd love to share it here.

Toy Master (Fullscreen Recommended)

Shortly thereafter this game failed hyper-casual marketing tests, I quit Kwest to start my new project under the code name: "Krypto Creatures", it's about generative NFT entities and related infrastructure. Involves AI, genetic programming, and blockchain technologies. Unfortunately, I didn't have enough resources to continue to work on it full-time so it's now my little personal weekends work.









Gallery

Here's more of my little personal weekends works. Shaders are an interesting way to demonstrate a high level of complexity (which I enjoy handling) and fascinating enough to attract your attention. While I don't consider myself an artist, I enjoy creating such things in a form of mathematical and engineering problem-solving. Besides, you might be interested in the infrastructure I've developed for this showcase. Click on arrow icons to navigate.









Resum-e










Contacts

GET IN TOUCH

e-mail: ravenametr@gmail.com

LinkedIn

GitHub