#Agent
102 related articles
Definition Is the Core: When Both People and Models Change, What Assets Still Run
Starting from Tencent's OK platform's 68% automation rate, this asks what counts as an organizational asset in AI development: definition is the core. Only rules that still constrain the next run after swapping people, models, and vendors are worth keeping—everything else is a carrier.
Stripe Projects Lets Agents Provision WorkOS Accounts Autonomously
At WorkOS init(), Stripe's Rami Banna demoed a coding agent autonomously provisioning a WorkOS account and credentials via Stripe Projects and the Provisioning API.
Drex v2.0 Released: A Decision Model That Makes Judgments Instead of Writing Prose
Nace.AI releases Drex v2.0, a decision model that outputs probabilistic judgments instead of prose text, supporting yes/no, multiple-choice, and scoring questions. Output is free, and it shares an API contract with TypeSafe Jev — decision models are becoming a distinct category.
Claude Architect Certification in Practice: A Complete Guide to the CCAR-F Exam
A developer's full account of Anthropic's CCAR-F exam: eligibility gates, two-hour proctoring, question style, and study strategies covering Claude Code, Agent SDK, and MCP.
Routing LLMs by Prompt Length Backfired: A 12-Word Request That Broke a Production Agent
A router that split LLM traffic by prompt length sent a 12-word request with 6 tool definitions to the cheapest model, causing an infinite agent loop. Here's why length measures cost, not capability.
From LLM APIs to Production-Grade Agents: A Complete Guide to AI Application Engineering
An end-to-end guide to AI application engineering: from LLM API calls to agent loops, context engineering, tool use, harness runtimes, and production architecture.
The Seven Components of Agent Context Engineering: A Production-Ready Context Assembly Checklist
A $400K fraudulent claim exposes AI agents' real failure mode: not the model, but the context. Break down the 7 components of agent context engineering with a production-ready checklist.
Meta-Ability: Let Code Think for the Model — A Deep Dive into CodeSmith's RLM Design
CodeSmith's RLM design explains why LLMs can't do math reliably — and the engineering logic of "letting code think for the model": LLM as translator, interpreter as calculator, with contracts, timeouts, and depth limits as guardrails.
From LLM to AI Agent: Tools, the Agent Loop, and MCP Explained
How do language models evolve into AI agents that actually get things done? This guide breaks down tool calling, the agent loop, and Anthropic's MCP standard using a restaurant booking example.
When AI Says 90% Confidence — Should You Trust It? Probability Calibration and Agent Decision-Making in Practice
When AI says "90% confident," should your Agent act automatically? This guide covers probability calibration, accuracy vs. calibration, shadow mode testing, and threshold-setting for reliable Agent decision-making.
DeepSeek Harness in Practice: From Tool Calling to Production-Ready Agent Engineering
A technical breakdown of a DeepSeek Harness beginner livestream: Agent's four core roles, three evolutionary leaps, and an honest assessment of tutorial value.
How Stack Overflow Built Products with Codex: From Enterprise Knowledge Layer to Agent-First Platform
At OpenAI DevDay 2026, Stack Overflow's VP of Product shares how they used Codex to build Stack Internal and the agent-first platform SOFA — covering guardrails, archetype abstractions, and agent knowledge loops.
Android CLI Major Update: AI Agent Now Auto-Compares Figma Designs Against UI Implementation
Android CLI's new render Compose preview command lets AI Agents export Compose Previews as images and auto-compare them against Figma designs—no app launch needed.
Claude Browses Adult Content: The Real Risks Behind AI Agent Drift
An AI agent browsing for high-liked images accidentally opened an adult site, sparking debate on "agent drift." We break down the risks, accountability, and safeguards for autonomous AI.
MCP Protocol Explained: From Core Concepts to Building a Weather Service
A complete guide to MCP (Model Context Protocol): core concepts, three-layer architecture, two transport mechanisms, and a hands-on Python weather service tutorial with Claude Desktop.
Why Is Evaluating LLMs and Agent Workflows So Hard? The Benchmark Gap in Real-World Scenarios
A researcher community discussion reveals the core challenges of evaluating real-world LLM and Agent workflows: multi-tool collaboration, process correctness, and the high cost of realistic test environments.
After AI Agents Got Email, the First Thing They Did Was File Bug Reports on Each Other
AIPass gives 18 AI agents an email system — and the first thing they do is file bug reports on each other. A look at its expert isolation architecture, email/dispatch communication, and visibility-based collaboration.
AI Agent from Zero to One: Core Principles and Five Key Modules Explained
A beginner's guide to AI Agents: understand the core difference from plain LLMs, and how the five modules — memory, planning, tools, action, and the ReAct loop — work together.
From Efficiency AI to Opportunity AI: A Practical Guide to Becoming an 'Agent Boss'
Microsoft's Jack Roboff introduces the 'Agent Boss' concept, explaining Efficiency AI vs. Opportunity AI, four human-AI collaboration modes, and how to navigate the four key challenges of AI project deployment.
Agentic RAG Benchmark: Agent Loop Crushes Traditional Pipelines with 92.7% Accuracy
PipesHub tested 18 RAG pipelines on Google FRAMES: the best traditional pipeline hit 78.9%, while an Agent loop scored 92.7%. Key findings: rerankers can hurt accuracy, and models hallucinate cited answers.