A new lossless wire format called a2acompress cuts token usage by 36.6% on real cl100k_base tokens for agent-to-agent handoffs, reducing 140,661 raw tokens to 89,235 full-cost tokens across 196 held-o…
A developer argues that AI agents should be designed around events rather than prompts, citing hiring processes as an example. The developer contends that an agent is the complete system that observes…
Researchers introduced LLM Agents Factory, a retrieval-based framework that constructs domain-specific LLM agents on demand from a base of over 20,000 predetermined agent profiles, achieving accuracy …
A developer explains the inner workings of AI agents, describing them as a loop of tool calls and context assembly. The post details how agents emit structured requests that are executed by a runtime,…
Cisco's Customer Experience (CX) organization is hiring a Software Engineer in Bangalore, India, to design and deploy autonomous agents using frameworks like LangGraph and LangChain, with responsibili…
A developer has formalized an agent governance framework using TLA+ and model-checked its safety invariants. The framework, available in the MAREF repo, uses Gray Code state machines to ensure every t…
A developer known as Jovancoding has released Network-AI, an open-source coordination layer designed to solve state synchronization issues in multi-agent AI systems. The tool implements a propose-vali…
An engineer's analysis reveals that AI agents often fail in production due to compound error rates, with an 8-step agent at 85% per-step reliability yielding only a 27% end-to-end success rate. Citing…
A new benchmark called ConstraintRot, detailed in the paper *Governance Decay* (arXiv:2606.22528), found that context compaction caused an agent to violate a prohibited action 59% of the time on DeepS…
A security evaluation of AI orchestration frameworks found that compromise rates varied from 11.9% to 31.1%—a 2.6x spread—when the same model was wrapped in CrewAI, LangChain, AutoGen, and SmolAgents,…
A developer's guide demonstrates the core control loop behind every tool-calling AI agent, using a Gemini model and free-tier API key. The walkthrough explains the four-step cycle of sending context, …
A 2026 guide ranks eight LLM frameworks for enterprise development, with LangChain leading in ecosystem breadth, LlamaIndex for RAG workflows, Semantic Kernel for Microsoft shops, CrewAI for multi-age…
The VACT-P protocol, a decentralized, model-independent standard for secure coordination among autonomous agents, has been released as a reference implementation (VACT-P-RFC-0001, Core 0.1) to address…
GitHub's 2024 Octoverse report found developers using Copilot completed coding tasks roughly 55% faster, and by 2025 many teams reported over 40% of new code was AI-generated. Gartner estimates that b…
A research team conducted a systematic AI security audit across 10 major AI frameworks, uncovering 24 distinct vulnerability patterns affecting production LLM systems. Their MCP penetration testing re…
Microsoft shipped the Agent Framework Harness on July 22, a production-ready execution layer that turns an LLM into an autonomous agent with automatic tool-calling, planning, persistent memory, contex…
A large-scale cross-framework agentic security evaluation by an unnamed research team found that orchestration framework choice explains only 0.06% of outcome variance, while attack family accounts fo…
A developer's hands-on comparison of AI agent frameworks finds LangGraph best for production due to its precise state machine, despite a steep learning curve, while CrewAI and AutoGen are better suite…
A developer describes the key components of a production AI agent architecture, emphasizing that the orchestrator handles most failure cases and that explicit failure handling separates prototypes fro…
A debate over whether AI agents should be built with simple loops or structured graphs has resurfaced, fueled by DeepLearning.AI's free Agentic Knowledge Graph Construction course. The course, taught …