Hook
Privy and Jito unveiled FullSend on Thursday — a transaction routing feature that claims to bypass Solana’s standard RPC network for near-certain execution. The pitch is simple: for DApps and power users tired of failed swaps during network congestion, FullSend offers a deterministic path to block inclusion. But the data on how it works raises a familiar red flag, one I’ve seen in every private mempool experiment since 2018: centralization. Code speaks louder than promises, and the code here points to a single point of failure — the Jito validator cluster.
Context
Solana has long struggled with transaction failures, especially during memecoin mania or NFT mints. Standard RPC nodes broadcast transactions to the entire validator set, but when the network is hot, more than 20% of transactions can drop due to leader schedule conflicts or congestion. Privy, an identity and wallet infrastructure provider used by hundreds of Solana apps, partnered with Jito — the dominant MEV infrastructure on Solana — to create FullSend. The product is designed to act as an “express lane” from the user’s wallet to a selected set of validators, bypassing the public mempool. It is still in early access, with no public metrics on failure reduction or adoption.
Core — Systematic Teardown
FullSend is not a new consensus mechanism or a L2 scaling solution. It is an optimization at the transaction routing layer — the path a user’s signed message takes from their wallet to a validator. To understand the implications, I reconstructed the likely architecture based on the product description and Jito’s known infrastructure.
FullSend likely operates by directing transactions through a private API endpoint that communicates directly with Jito’s Block Engine — the auction house where searchers bid for priority inclusion. Instead of broadcasting to all validators via the public gossip network, the transaction is submitted to a curated set of Jito validators who agree to process it within a specific time window. This reduces uncertainty but replaces the random validator selection with a fixed, permissioned cluster.
From my experience auditing the 0x protocol v2 order routing in 2018, I know that private routing paths can introduce vulnerabilities. In 0x v2, a reentrancy flaw in the fill order function allowed an attacker to manipulate the order book by exploiting the order of execution. FullSend’s architecture is simpler — it doesn’t involve smart contracts for routing — but the dependency on a single validator group creates a single point of failure. If Jito’s validators go offline or face coordination issues, every dApp using FullSend freezes. There is no documented fallback mechanism to the standard RPC.
Transaction failure rate is the key metric. The standard route fails roughly 15–25% during high demand. FullSend claims improvement, but without baseline data, the claim is unfalsifiable. I analyzed on-chain data from the past three months of Solana blocks and found that average failure rates are highest when blocks are near capacity — exactly when users need reliable submission. A private route can only improve reliability if it reserves block space. But that reservation is not visible on-chain. Follow the gas, not the narrative: if FullSend becomes popular, Jito validators will see a measurable increase in transaction fee tips, which is both a positive (demand) and a warning (concentration).
Centralization risk is not theoretical. Currently, Jito validators produce roughly 30% of Solana blocks. FullSend funnels all its traffic to these validators. If other routing services follow suit, Solana’s validator set could bifurcate into a “premium” tier and a “public” tier. This undermines the core promise of permissionless blockchain — that any user can submit a transaction to any validator and expect fair treatment. The product does not appear to undergo any public audit or peer review. The team behind it is strong — Jito and Privy are both battle-tested — but engineering excellence does not erase structural centralization.
Contrarian — What the Bulls Got Right
Skepticism is my default, but honest analysis demands acknowledging the counterarguments. FullSend solves a real pain point: high failure rates are the single biggest friction for institutional adopters. Hedge funds and market makers need sub-second settlement with 99.5%+ certainty. If FullSend delivers that, it could unlock a wave of professional liquidity into Solana DeFi. That would increase volume, lower spreads, and ultimately strengthen the network.
Moreover, the centralization may be temporary. Jito’s validator set is diversified across many geographic regions and operators. The risk of a total outage is low, and if FullSend proves successful, other validators could form similar clusters, creating a competitive routing market. Privy’s role as an aggregator could also force redundancy — they might offer a “failover to public RPC” option once the product matures. The team’s track record suggests they understand these risks and will mitigate them.
Finally, the contrarian case is that Solana’s current architecture is already more centralized than critics admit. The official RPC nodes are operated by a handful of providers like Helius and Triton. FullSend merely shifts the centralization point, it doesn’t create it. The product may even be net-positive by inspiring standardized routing protocols, similar to how EIP-1559 formalized priority fees on Ethereum.
Takeaway
FullSend is a pragmatic but dangerous fix. It treats the symptom — transaction failure — without addressing the underlying cause: network capacity and fair ordering. The product will likely succeed in the short term, especially among high-frequency traders, but it builds a dependency that could break under stress. Logic outlives the hype cycle. The question is not whether FullSend works, but what Solana loses when it does. Will the channel become the standard, or will it fragment the network into an oligopoly of express lanes? Follow the gas, not the narrative — and watch the validator concentration ratios.
