DeFi engineering part 2: Four challenges
How Efficient Frontier’s DeFi team moves fast without breaking things. Overview of the four main challenges faced by technical teams interfacing with the DeFi ecosystem
1. Blockchain engineering challenges for financial companies
While Ethereum is the dominant network in the DeFi world, there are already over 50 different chains operating in this sphere trying to compete. Understanding which blockchains to connect to and then the technical integration to each of the chains presents a challenge. Each chain requires understanding of its capabilities, integration to its nodes, exploring costs, latency, players and economics. To be on the cutting-edge we need access to the fastest RPCs, private nodes and mempool explorers.
Do you think Ethereum will continue to lose market share? Which network do you think poses the biggest threat to Ethereum?
2. Smart Contract Challenges
Given Ethereum’s dominance in the market, most of the chains have built their platforms to support EVM (Ethereum Virtual Machine). This allows compatibility to popular developer tools, interfaces and programming languages. EVM contracts are written in Solidity or Viper — code which can be easily migrated, analyzed, reused and extended.
However, EVMs are far from perfect, and many competitors are innovating in order to address prevalent issues such as latency or cost. WebAssembly has become more popular; Rust is a favored choice for Solana; and StarkNet invented Cairo language for provable programs.
This challenge demands deciding how much time to allocate to each protocol, keeping up to date on new innovations, keeping an open mind and being agile.
3. R&D Challenges
The key to success in R&D is automation. Our team is small but fully autonomous and has all the tools and skills to analyze, design, build, test and release features fast. We manage our code in a monorepo, cover it with unit and integration tests, use github actions for CI/CD and monitor metrics with Kibana and Grafana. The combination of Tenderly and Truffle Suite creates different snapshots of blockchain state, and allows us to simulate our algorithms behavior and ensures functionality with regression tests. This leaves us time to focus on business logic and release features without friction.
In order to scale fast without breaking things, we designed microservices architecture and heavily rely on EKS (Amazon Managed Kubernetes) with autoscaler, helm charts and terraform deployment. To support the whole operation, monitor performance, cost and health, build dashboards, configure alerts and maintain nodes without downtime, we must continue to develop our devops muscle.
4. Security Challenges
Interacting with many platforms and smart contracts exposes us to critical security challenges. We read and understand smart contracts’ code, we analyze security audits provided, explore code and economics vulnerabilities, verify, speak to and help projects.
That said, nothing can guarantee 100% safety. History shows that even the best audited protocols may have bugs and risks to be exploited. That’s why we also take active prevention measures — monitoring and analyzing onchain data (Ethereum alone produces millions of transactions daily), setting smart alerts and being design partners with startups that specialize in that field.
Efficient Frontier activity results in cumulative crypto trading volume of billions of dollars and a security breach could cause huge losses. We use industry standard platforms such as Gnosis, Copper and Fireblocks to safely store our assets in vaults, however our daily operation requires immediate sub-second access to the funds. I can’t disclose technical details due to its sensitivity, but optimization of DeFi operations critical paths without compromising on security is one of the toughest challenges I met in my life.
What do you think are the biggest challenges for DeFi? Share it with us so WAGMI
If this got you excited about DeFi engineering and you want to join our journey, we have lots of open positions
