Tag
ai
4 posts
- A Claude-centric software architectureFor thirty years, software has meant humans translating changing requirements into fixed code. Once a runtime like Claude Code is strong enough, the more efficient approach is to separate the two: settle deterministic logic into a data layer, a function layer and Skill methodology, and hand the uncertain requirements to the runtime to understand and render in real time. With quant-researcher as a worked example.
- Three things every AI agent engineer should know: agent architecture, prompt engineering, and MCPA platform-agnostic guide: the agentic loop and multi-agent orchestration, context passing and error propagation, few-shot and structured output, controlling false positives, and the structure, message format and a minimal server implementation of the Model Context Protocol.
- Goodbye Photoshop? Ten everyday image edits done with Gemini Nano-Banana promptsForget layers and tool palettes. Ten common image-editing jobs, from cutting out a background to style transfer and outpainting, each with the traditional Photoshop workflow and the single Nano-Banana prompt that replaces it.
- Google's official prompting guide for Nano-Banana, annotatedWhat Gemini 2.5 Flash Image (Nano-Banana) can do, the one principle behind good prompts (describe the scene, don't list keywords), and Google's example prompts for photorealistic scenes, stickers, logos, product shots, editing, inpainting, style transfer and multi-image composition.