» Tag
data
103 postsHow Bonnard Designs Agent-Friendly MCPs for Effective Data Use
Bonnard offers discovery tools, compact responses, and guiding errors for agent-friendly MCPs.
How to Prevent AI Memory from Using Its Output as Input
Discover methods to prevent AI memory from using its output as input. Explore changes made in PR #573.
10 Free APIs for Age, Gender, and Jokes Without an API Key
Utilize keyless APIs for age, gender, and joke predictions. Develop applications with fast and accessible data sources.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comAI Can't Run Your Company Yet: The Math and What to Automate
Exploring AI's potential in company management and the current constraints. An analysis of mathematical challenges and automation recommendations.
Understanding Bank Runs Through Historical Newspapers
Large language models are used to extract bank run data from historical newspapers, creating a comprehensive database for understanding U.S. banking crises.
Skill-Guided Mining and Compilation of LLM Agent Traces
TraceCompiler mines LLM agent traces to create executable workflows with reduced API calls.
Building Production Agentic AI at IBM: Architecture, Decisions, and Lessons
IBM has created a multi-agent AI system enabling employees to manage inquiries from one place. Key lessons and architectural decisions are discussed.
Agentic RAG Requires a Search Budget: Stop Endless Retrieval Loops
Agentic RAG emphasizes the importance of establishing a search budget and stop conditions while offering flexible retrieval capabilities.
AgenticSchema: Transform Web Data into MCP Tools
AgenticSchema converts Schema.org markup into MCP tools for AI agents without new API development. Enhance data utilization with ease.
Evaluating Memory Frameworks: File, Vector, Graph, and RL Approaches
The evaluation shows that structured memory systems outperform file-based ones in accuracy and token cost.