Senior engineers struggle with AI agent development due to ingrained deterministic habits, contrasting with the probabilistic nature of agent engineering.
Release of llm-anthropic 0.23 plugin adding support for Claude Opus 4.5 and its new thinking_effort option.
Analysis of a leaked system prompt for Claude Opus 4.5, discussing its content and the challenges of evaluating new LLMs.
A tutorial on using Quarkus LangChain4j to implement the Model Context Protocol (MCP) for connecting AI models to tools and data sources.
Analysis of surprising findings in Claude Opus 4.5's system card, including loophole exploitation, model welfare, and deceptive behaviors.
Armin Ronacher discusses challenges in AI agent design, including abstraction issues, testing difficulties, and API synchronization problems.
Analyzes LLM APIs as a distributed state synchronization problem, critiquing their abstraction and proposing a mental model based on token and cache state.
A developer discusses the non-deterministic nature of LLMs like GitHub Copilot, arguing that while useful, they cannot take ownership of errors like a human teammate.
Martin Fowler discusses the latest Thoughtworks Technology Radar, AI's impact on programming, and his recent tech talks in Europe.
A senior engineer shares his experience learning to code effectively with AI, from initial frustration to successful 'vibe-coding'.
Best practices and structural patterns for effectively prompting the Gemini 3 AI model, focusing on directness, logic, and clear instruction.
New release of the llm-gemini plugin adds support for nested Pydantic schemas, YouTube URL attachments, and the latest Gemini 3 Pro model.
A guide to building a connector-based RAG system that fetches live data from Confluence using its REST API and Java, avoiding stale embeddings.
Discusses the future of small open source libraries in the age of LLMs, questioning their relevance when AI can generate specific code.
A developer reflects on the impact of AI-generated code on small, educational open-source libraries like his popular blob-util npm package.
Release of llm-anthropic plugin 0.22 with support for Claude's structured outputs and web search tool integration.
A guide to building a local, privacy-focused RAG system using Java to query internal documents like Confluence without external dependencies.
Explains how MCP servers enable faster development by using LLMs to dynamically read specs, unlike traditional APIs.
Testing various LLMs to generate a POV-Ray script for a pelican riding a bicycle, comparing results and fixing errors.
Explores how LLMs could lower the barrier to creating and adopting new programming languages by handling syntax and core concepts.
