Loading video...
Video Failed to Load
Arena Conversations: Peter Gostev sat down with Matt – Trigger.dev, founder and CEO of Trigger.dev, to discuss building infrastructure for long-running agents, including retries, API reliability, and why customers are switching frontier models faster than ever. Watch the full episode at the link below.
16,619 views • 25 days ago •via X (Twitter)
5 Comments

@petergostev @mattaitken @triggerdotdev Retries and API reliability are the foundation for long-running agents: without durable recovery, model switching just amplifies operational fragility. The conversation’s focus on production infrastructure is a useful complement to benchmark-driven discussions.

@petergostev @mattaitken @triggerdotdev Long-running agents need retries and observability as much as stronger model intelligence.

@petergostev @mattaitken @triggerdotdev Great talk as always. This brings a lot of value. You and @88krxxs are my go to accounts.

@petergostev @mattaitken @triggerdotdev Long-running AI agents are only as reliable as the infrastructure behind them. The retries and API reliability discussion is especially interesting as model switching becomes the norm. 🔥

@petergostev @mattaitken @triggerdotdev Retries and reliability first. Switching models is cheap if the workflow stays.


