Tagged: agents
Showing 1–5 of 32 articles
Agent Session Logs Already Know What You Spent I summed a month of agent spend straight from local session logs. Per message, per subagent, exact to the cent. No metering needed. Read article How Many Companies Know Their Token Bill? AI cost tracking tools are appearing fast. Metering was never the hard part; knowing which piece of work owns the spend is. Read article Four Runtimes Built the Same Memory Ladder Headlong, OpenViking and ai-coworkers all landed on layered memory. Hermes built a plugin interface instead. What each one refused to build is the interesting part. Read article Shipping Headlong's Design Doc: The Memory Ladder in ai-coworkers Laude published a progressive-resolution memory design and marked it unimplemented. This is what it looks like shipped in another runtime, and what I refused to take. Read article Headlong vs ai-coworkers: What Should an Agent Do When Nothing Is Happening? Laude's Headlong thinks continuously and costs $1-2 an hour at idle. My runtime returns before the model is called. One design question separates them. Read article