Author: Oscar Serra
-
Total Recall: Pointer-Based Compaction and Task-Conditioned Retrieval for Persistent LLM Agents
Most AI agents compact old context by summarizing it — quietly losing the one detail that mattered. Total Recall is a lossless, event-sourced memory architecture that evicts with recoverable pointers instead of summaries, and keeps memory bounded. Read more
-
Running an AI Agent 24/7: What Nobody Tells You
75 crashes, 45 wrong messages, and 138 restart loops. What actually happens when you leave an AI agent running around the clock. Read more
-
How to Track AI Token Costs in Real Time
See where every AI token goes — in real time, not on next month’s invoice. Practical strategies that cut our daily costs by 60%. Read more
-
How to Set Up Self-Improving AI Cron Jobs
The META pattern: AI cron jobs that rewrite their own prompts after each run. Day 1 mediocre, Day 30 expert. No human intervention. Read more
-
The Field Guide for AI Agents: 32 Lessons from Running One 24/7
32 hard-won lessons from running an AI agent 24/7. The operational manual nobody else has written. Read more
-
TinkerClaw: We Forked OpenClaw. Here’s What We Changed and Why.
262 fork commits, a real-time token dashboard, cognitive memory, and self-improving crons. What TinkerClaw adds to OpenClaw — and why we built it. Read more
-
We Taught Our AI to Teach Another AI
Two OpenClaw instances, same family, different operators. How we transferred 32 operational lessons from one AI agent to another — and wrote the field guide for everyone else. Read more
-
What Running an AI Agent 24/7 Actually Costs
After 6 weeks of running AI agents around the clock, we learned where every token goes — and wrote it all down. Real numbers, real crashes, real lessons. Read more