Today Anthropic’s Zach Witten takes us on a deep dive into Anthropic’s cutting-edge AI models—Claude Haiku, Sonnet, and Opus—exploring their safety-first approach to generative AI and sharing essential tips for prompt engineering.Topics Include:Introductions, about Anthropic3 models: Haiku, Sonnet and OpusScaling laws for hardware, data and computeCompeting to be safest AI solutions, safety-first organizationLeader in jailbreak resistanceInterpretability features and breakthroughs for AI modelsBasics of prompt engineeringImproving prompts with ClaudeDetails matter – small changes to spelling, context will greatly improve resultsSystem prompt – role setting will improve results (i.e. “You are an expert mathematician…” for math queryBe clear and direct – use XML tags where possibleEncourage Claude to think step-by-step – answering fast comes with accuracy riskUse examples to provide additional clarity to ClaudeBonus tips for image-based prompt engineeringQ&A 1) Who wrote the meta-prompts in the cookbook?Q&A 2) Guidance for writing prompts for prompt generatorQ&A 3) Best practices for tabular and structured dataQ&A 4) Maintaining “tone” across hundreds/thousands of responsesQ&A 5) Reverse engineering a prompt

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Transcript

Related