Agent Design Is Still Hard
I felt like it might be a good time to write about some new things I’ve learned. Most of this is going to be about building agents, with a little bit about using agentic coding tools. TL;DR: Building agents is still messy. SDK abstractions break once you hit real tool use. Caching works better when you manage it yourself, but differs between models. Reinforcement ends up doing more heavy lifting t
MCPサーãƒãƒ¼ã‚’使ã†ãªã‚‰ Prompt Caching ãŒå¤§åˆ‡ã ã¨æ€ã„知ã£ãŸè©±
MCPサーãƒãƒ¼ã¨é€£æºã™ã‚‹ã‚¢ãƒ—リ使ã£ãŸã‚‰ã€åˆ©ç”¨æ–™ãŒé«˜ã‹ã£ãŸ MCPサーãƒãƒ¼ã¨é€£æºã™ã‚‹ãƒãƒ£ãƒƒãƒˆã‚¢ãƒ—リを作ã£ã¦ã„ãŸã®ã§ã™ãŒã€ã€Œé–‹ç™ºä¸ã§ã»ã¼è‡ªåˆ†ã—ã‹ä½¿ã£ã¦ãªã„ã®ã«ç”ŸæˆAIã®åˆ©ç”¨é‡ãŒã‚„ãŸã‚‰é«˜ãã？ã€ã¨æ€ã„ã¾ã—ãŸã€‚ Claudeã®MaxプランãŒæœˆé¡100ドルã§ä½¿ã„放題ã§ã‚ã‚‹ã“ã¨ã‚’考ãˆã‚‹ã¨ã€é–‹ç™ºä¸ã«è»½ã使ã£ãŸã ã‘ã§æ•°åƒå††ã‹ã‹ã‚‹ã‚¢ãƒ—リãªã‚“ã§ã‚³ã‚¹ãƒ‘悪ã™ãŽã¦ã€ã¾ã¨ã‚‚ã«ä½¿ãˆãªã„ã§ã™ã。 調ã¹ã¦ã¿ãŸ ãªãœãã‚“ãªã«é«˜ã„ã®ã‹åˆ†ã‹ã‚‰ãªã‹ã£ãŸã®ã§ã€Claudeã®ãƒã‚°ã‚’å–ã£ã¦èª¿ã¹ã¦ã¿ã¾ã—ãŸã€‚ ãƒã‚°ã®å–ã‚Šæ–¹ã¯å…ˆæ—¥æ›¸ã„ãŸè¨˜äº‹ã‚’å‚ç…§ãã ã•ã„。 ãã—ã¦ã€å–å¾—ã—ãŸãƒã‚°ã‚’確èªã—ãŸã¨ã“ã‚ã€ã€ã€ {"input_tokens":21786,"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"output_tokens":85}} "input_tokens":21786
Effectively use prompt caching on Amazon Bedrock | Amazon Web Services
AWS Machine Learning Blog Effectively use prompt caching on Amazon Bedrock Prompt caching, now generally available on Amazon Bedrock with Anthropic’s Claude 3.5 Haiku and Claude 3.7 Sonnet, along with Nova Micro, Nova Lite, and Nova Pro models, lowers response latency by up to 85% and reduces costs up to 90% by caching frequently used prompts across multiple API calls. With prompt caching, you can
1
メンテナンス
ランã‚ング
ãŠçŸ¥ã‚‰ã›
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
処ç†ã‚’実行ä¸ã§ã™
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Fix over-caching with Dynamic IO caching in Next.js 15 - LogRocket Blog
Gitfaster
{{#tags}}- {{label}}
{{/tags}}