Since the beginning of Proxikle, our vision has been clear to build meaningful, scalable and resilient products for Web3 users and developers. We’ve always believed that the best products are those built with intent, from the ground up, not forked or borrowed. Over time, many of you who use our products may have noticed that our team isn’t confined to one geography. Much like blockchain nodes, we’re distributed across North America and Asia, working asynchronously and united by the shared purpose to engineer products that truly matter.
With our team distributed across continents, where does our infrastructure live?
Unlike traditional Web3 hyperscalers who rely heavily on centralized cloud providers like AWS, Azure or Google Cloud, our blockchain infrastructure is not hosted on any of them. Instead, we run dozens of nodes and modified databases on bare-metal machines that we have rented and configured ourselves, all running behind custom load balancers. This setup didn’t come easy. It took us months of research, testing and experimentation to understand how to handle ultra-large blockchain datasets efficiently without relying on AWS or Google Cloud. We’ve spent countless hours on R&D, building custom solutions on top of battle-tested open-source software to shape the infrastructure that powers Proxikle, which users will learn more about in the coming months.
But it was worth it. Because now, we have an infrastructure that can scale, can run independently, and gives us complete control not just over the code but the data itself.
You might wonder why we choose to build everything from the ground up instead of relying on readily available solutions even if that means extra effort and cost.
At Proxikle, we have made a deliberate choice not to depend on pre-built Web3 frameworks or third-party tools. Everything we create from the data pipelines that power our products to the UX design and translation workflows is developed entirely in-house. We have also never used LLMs to design interfaces or translate content. This decision keeps us close to our target audience, helping us understand their behavior and preferences at a much deeper level. This approach may be harder and slower, but it gives us the freedom to engineer products exactly the way we envision them, without the limitations or compromises that come with using existing Web3 solutions.
We have all seen how many Web3 projects suffer from poor user experience and fragile engineering often by forking an existing codebase, raising capital, and eventually fading away within months. Our team learned from those patterns early on. To manage our products efficiently at scale, we chose not to depend on third-party APIs or shared blockchain data. Instead, we built our own APIs by running our own blockchain nodes on bare-metal servers that we control. This strategy has not only improved performance and reliability but also drastically reduced our operational costs allowing us to scale sustainably while maintaining full ownership of our engineering model.
Currently, there is a minor reliance on AWS, used only for hosting client source code and customer data. This dependency will be phased out once we begin raising our rounds of funding and expand our own infrastructure capacity. The database that powers our blockchain querying layer is custom-built on PostgreSQL, enhanced with internal modifications for handling blockchain-scale data. On top of that, we have built custom indexers from scratch in Rust, optimized for speed, scalability, and fault tolerance.
Our infrastructure stack primarily relies on Rust & Java, the core language used across our systems from massive indexers and data listeners to the database that sustain Proxikle’s ecosystem.
Building yet another wrapper on AWS, Google Cloud or using Web3 APIs to build and ship a blockchain Explorer may be quick and convenient but rarely creates real value for end users. It’s easy to ship a product that looks functional on the surface, but if it’s built entirely on someone else’s infrastructure, the control and innovation remain limited. At Proxikle, we choose a different path. By engineering and operating our own blockchain nodes and managing the entire data layer ourselves, we’ve built something deeper that is a decentralized cloud platform powered by our own infrastructure.
Owning both the nodes and the data gives us unmatched control, reliability, and flexibility in what we build. This approach not only strengthens our engineering foundation but also aligns perfectly with our long-term vision for a truly decentralized Web3 ecosystem.
Most of today's cloud services are built on top of existing hyperscale cloud infrastructure provided by AWS, Azure or Google Cloud. When you build a “Cloud platform” on top of another cloud, you inevitably inherit its limitations, design choices, and operational model. The same applies to Web3. If you are using third party’s APIs to launch a blockchain explorer or analytics products, you’re essentially building a UI shell on top of the third party’s infrastructure. In that setup, it’s the provider not you who holds the real control and earns the majority of the revenue.
With that said, platforms like Netflix or Coinbase aren’t trying to become infrastructure providers themselves. For them, building on AWS makes perfect sense because it allows them to launch products faster and focus on their core offerings. At Proxikle, however, our mission goes beyond building products on the cloud. We’re building the infrastructure itself, the decentralized foundation that will power the next generation of Web3 applications.
.png){kind=small}
.png){kind=small}
