SRSai Ram Kothainsairamkotha.hashnode.dev·10h ago · 8 min readReasoning Effort Across AI Providers: One Knob, Four MeaningsEvery major AI provider now lets you turn reasoning up or down. The names look alike: effort, thinking level, thinking budget. The behavior is not alike. Some settings are a hint, some are a hard cap,00
SRSai Ram Kothainsairamkotha.hashnode.dev·15h ago · 9 min readHow AI Chat "Compaction" Works: Summaries That Shrink the ContextA long AI chat gets heavier with every message, because the whole conversation is sent again each time. Compaction is the fix: replace the old part of the conversation with a short summary and carry o01A
SRSai Ram Kothainsairamkotha.hashnode.dev·15h ago · 6 min readCost Per Task: What a $0.10 Subagent Changes for AI AgentsOn October 7, Anthropic released Claude Haiku 5.5, its smallest model, at $0.10 per million input tokens and $0.50 per million output tokens. That is a headline price cut. The more useful question for00
SRSai Ram Kothainsairamkotha.hashnode.dev·15h ago · 8 min readWho Gets Access? A Week of Gated, Open and Unreleased AIThree big AI announcements landed within a day of each other this week. They look unrelated: a batch of math papers, a cybersecurity access program, and a new open model. But they answer the same ques00
SRSai Ram Kothainsairamkotha.hashnode.dev·15h ago · 10 min readWhat Happens When You Press Send: An AI Chat, From Basic to AdvancedUpdate, October 9, 2026: Anthropic cut the Sonnet 5.5 cache-read price from $0.20 to $0.10 per million tokens on October 7-8. I have updated the worked example in Level 5 to the new price (about $0.2400
