We built not for the peak, but for the valley. Yet in the current bear market, the valley is precisely where we must scrutinize the narratives that promise to lift us out. The parallel EVM race is not just a technical contest; it is a litmus test for the soul of crypto infrastructure. Over the past month, I have audited the whitepapers and technical documents of Monad and Sei, and I returned to a cabin in Yilan—the same one from 2022—to reflect on what we are actually building.
Context: The Performance Mirage
Parallel EVM emerged as a solution to Ethereum's sequential execution bottleneck, promising to unlock thousands of TPS by executing non-conflicting transactions simultaneously. Monad, with its roots in Jump Crypto, touts a “physical parallel” architecture: true parallel execution combined with a custom state database (Monad Db) and a Byzantine fault-tolerant consensus (MonadBFT). Sei, on the other hand, deployed an “optimistic parallel” approach in v2—executing transactions first and reverting conflicts later. The market has already priced Sei at a fully diluted valuation of $130 billion, with a TVL of only $1.8 billion, yielding a staggering FDV/TVL ratio of 72x. Monad, still pre-token, has an OTC valuation of $3–5 billion, with no mainnet yet.
Core: The Technical Reality Behind the Hype
Let me be clear: both approaches have merit, but neither is a panacea. During my 2017 audit of OmniChain, I learned that the most elegant whitepaper can hide fatal flaws in tokenomics. Today, the same principle applies to execution models. Monad’s physical parallelism requires a fundamental redesign of Ethereum’s state model. The team claims 10,000 TPS, but I have seen similar claims in the 2021 L1 hype cycle—Solana, Avalanche, Fantom—all of which achieved far less under real-world conditions. The bottleneck is not CPU cores but I/O contention. Even with optimized storage, read-write conflicts will throttle throughput when the network reaches scale. Based on my experience designing governance frameworks for DAOs, I know that the most complex systems often fail at the simplest interfaces: the data layer.
Sei’s optimistic parallelism is technically less ambitious, but it has a practical advantage: it works today. The Sei mainnet supports parallel EVM, though with a caveat: the optimistic approach relies on deterministic re-execution, which adds latency and can lead to wasted computation. The network’s 210 million cross-chain active addresses and 150+ projects suggest a growing ecosystem, but I question the quality of those users. Many are likely incentivized by farm-and-dump mechanisms. The real test is retention, not vanity metrics.
The true risk lies in the narrative. Sei has successfully branded itself as “the parallel EVM chain,” but this branding is a double-edged sword. When the hype fades—and it will, in a bear market—the market will demand real utility. A $130 billion FDV for a chain with $1.8 billion TVL is unsustainable. Trust is the only protocol that cannot be coded. And trust, in this case, means believing that Sei can sustain its valuation without a massive influx of new users.
Contrarian: The Silent Flaw of Parallel Execution
Here is the counter-intuitive truth: parallel EVM may not actually solve the most critical problem for developers. The majority of smart contracts involve interdependent state changes—DeFi protocols, for instance, rely on sequential execution to maintain atomicity. Parallel execution works best for independent transactions, like simple token transfers, but not for complex DeFi operations. The market is betting on a technology that largely benefits use cases like NFT minting and DEX swaps, which are already handled by L2s like Arbitrum and Optimism. The real need is not more throughput; it is better composability and lower latency for cross-chain interactions.
Moreover, the competition is not just between Monad and Sei. Neon, Artela, and even Ethereum’s own L2s are exploring parallel execution. The winner will not be the chain with the highest TPS, but the one that offers the most developer-friendly environment—and that means robust tooling, low fees, and a strong community. We don’t need more users; we need more stewards.
Takeaway: The Valley Will Decide
In 2024, I founded The Alignment Circle to mentor builders on ethical governance. I learned that sustainable protocols are those that survive the valley, not the peak. Monad has the technology, but it has no mainnet. Sei has the users, but it has a valuation that defies gravity. The next 12–18 months will be the “mainnet season” for parallel EVM. When the bear market deepens, those who built for the chart will fade, and those who built for the soul will endure. The question is not who has the fastest chain, but who has the most resilient community.
We built not for the peak, but for the valley. The valley is coming. Let us see who is still standing.