Agent Memory vs RAG vs Long Context: Which One You Need (With Costs)
11 min read
RAG, agent memory or just a bigger context window? Cost per query at October 2026 prices, a runnable cost model, and when plain docs beat all three.
#system-design
2 posts
LLM API Budget Caps: Circuit Breakers, Not Dashboards
24 min read
Hard LLM API budget caps that stop spend before the call: reserve-then-settle in Redis, per-tenant quotas, a spend circuit breaker and a kill switch.
