TRENCHNET uses Jev to pre-filter wallet events before expensive analysis
Jev first classifies each on-chain event as skip, flag or investigate, and only calls Astra when deeper analysis is needed, which the author says cuts the agent team's yearly cost from $19,000 to $500.
Diogo Almeida, founder of TypeSafe and former OpenAI researcher: Jev Engineering. Imagine your entire agent team costs $500 a year. You used to budget $19,000 for the same workload. The savings are simple: Jev from TypeSafe handles routine decisions. The expensive model only gets called when a deeper analysis is needed. Here’s how I divide the roles in TRENCHNET: > Jev decides: skip, flag or investigate. > The graph checks which wallets regularly move together. > Astra handles the deeper analysis. > Telegram delivers the alert with transaction links. A wallet buys. Another follows. Both buy again. Each event could trigger an expensive analysis. Jev first determines whether one is needed at all. Several matching buys look convincing. Until the history shows the same group repeating a familiar pattern. You keep the final decision. Astra gets the difficult work. The market never closes. But the expensive model doesn’t need to discuss every transaction all night. I break down how this works in TRENCHNET in the post below ↓
This post quotes another post: “Using Jev as a routing layer for pump.fun trade monitoring”