Tag: #mcp
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 38 posts
Tool Surface Design — One Schema Line Moves the Success Rate
Adding more tools and watching the agent success rate drop is not rare. The tool surface is the agent interface, and the names, descriptions, parameters, failure returns, and response sizes are all design material. Part
2026-08-12 · 5 min read #llm#agent#harness-engineering#하네스엔지니어링#AI에이전트Running AI Agents in Production — Idempotency, Budgets, and Confidently Wrong Answers
Moving an agent from prototype to production exposes an operational surface you only discover late: idempotency for retried tool calls, budgets and step limits, observability over non-deterministic control flow, per-tool
2026-07-31 · 16 min read #ai#agents#observability#reliability#mcpHow AI Agents Fail in Production — 14 Failure Modes, and Why Retrying Is Not Safe
Putting agents in production hurts in three places. First, failures come from system design, not the model — UC Berkeley's MAST study classified 1642 execution traces into 14 failure modes, and 44.2% of them were system-
2026-07-17 · 20 min read #ai#agents#observability#reliability#mcpMCP Drops Sessions — Reading the Stateless Core in the 2026-07-28 Revision
The next MCP specification revision, 2026-07-28, is the biggest change since launch. Its core move is stripping state out of the protocol layer — the initialize handshake and Mcp-Session-Id sessions disappear, and every
2026-07-16 · 16 min read #mcp#ai#protocol#agents#integrationThe Model Context Protocol (MCP): An Engineer's Reference
The Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to LLMs. This is an engineer's reference grounded in the official docs: why it solves the M×N integration problem, t
2026-07-11 · 6 min read #mcp#ai#agents#llm#protocolThe Complete Guide to AI Agents in 2026 — From Coding to Everyday Life: Claude Code, MCP, Work Automation, and Multi-Agent Orchestration
In 2026, AI agents are everyday tools, not demos. This guide covers what an agent actually is (LLM + tools + loop), workflows versus agents, Claude Code in depth (CLAUDE.md, Skills, Hooks, parallel subagents), when to re
2026-07-02 · 36 min read #ai-agents#claude-code#mcp#automation#multi-agent2026 IAM Trends — AI Agent Identity, MCP Authentication, Verifiable Credentials
The 2026 IAM landscape is being reshaped by passwordless-by-default, an explosion of non-human identities in the form of AI agents, and the OAuth 2.1-based standardization of MCP authorization. From on-behalf-of delegati
2026-06-12 · 16 min read #iam#oauth2#mcp#ai-agent#keycloakBuilding MCP Servers in Practice — How to Connect Your Tools to Every AI Agent
A hands-on guide to building MCP (Model Context Protocol) servers, the de facto standard of 2026. We implement an internal wiki search server in TypeScript and Python, and cover tool design principles, OAuth 2.1 authenti
2026-06-12 · 17 min read #mcp#ai-agent#typescript#python#oauth2AI Agent Frameworks 2026 Deep Dive — LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen, PydanticAI, Mastra, DSPy, MCP Complete Guide
A complete picture of the AI agent framework market as of May 2026. LangChain holding share despite brand fatigue, LangGraph state-graph pattern, LlamaIndex RAG depth, CrewAI/AutoGen multi-agent, PydanticAI type safety,
2026-05-16 · 18 min read #ai-agent#langchain#langgraph#llamaindex#crewaiAI Agent Frameworks in 2026 — A Deep Dive on LangGraph, AutoGen, CrewAI, OpenAI Agents SDK, Anthropic Agent SDK, and More
In 2026 the AI agent framework landscape has exploded into a dozen camps. Vendor SDKs from OpenAI and Anthropic, orchestration frameworks like LangGraph and AutoGen and CrewAI, minimalist contenders like smolagents and M
2026-05-15 · 19 min read #ai-agent#langgraph#autogen#crewai#smolagentsBuilding an AI Teammate with a Slack Bot — Wiring Up Claude, Gemini, OpenClaw + Extending Tools with MCP (2026 Hands-On)
The highest-leverage surface for deploying AI isn't the IDE — it's Slack, where the whole team uses it, right where the work happens. This is a follow-along hands-on. Create a Slack app, stand up a minimal bot that respo
2026-05-14 · 16 min read #slack-bot#llm#claude#gemini#openclawBuild Your Own Claude Code Skill — A Hands-On Deep Dive Into SKILL.md, Invocation, and Distribution (2026, english)
Agent Skills were introduced by Anthropic in October 2025 and opened as a standard in December 2025. As of May 2026 they are the default way to extend Claude Code. The mechanism is small — a SKILL.md with YAML frontmatte
2026-05-14 · 23 min read #claude-code#skills#anthropic#ai-tooling#hands-onAI Coding Workflow Best Practices in 2026 — CLAUDE.md, AGENTS.md, .cursorrules, Skills, Subagents, MCP Deep Dive
The standards AI coding tools are converging on in 2026 — CLAUDE.md, AGENTS.md, .cursorrules, .github/copilot-instructions.md, plus Claude Code Skills, subagents, hooks, MCP, and permission modes. What goes in each file,
2026-05-14 · 20 min read #ai-coding#claude-md#agents-md#cursorrules#claude-codeThe Complete Guide to AI Development Automation — GitHub Integration, Ticket-Based Agentic Workflows, Copilot, Claude Code, Devin, Jules (2025)
2025 was the year AI moved from 'suggesting code' to 'owning the ticket.' Assign a GitHub Issue to an AI agent and it rolls forward on its own — branch creation, implementation, PR, passing CI. A 4-level maturity model,
2026-05-14 · 29 min read #ai-development#automation#github#copilot#claude-codeGemini CLI Practical Guide: How Developers Should Decide Whether to Adopt a Terminal-First AI Agent
A practical 2026 guide to Gemini CLI for developers who want to understand where it fits versus IDE-first tools, how plan mode, hooks, MCP, and scripting work, and how to roll it out safely on a team.
2026-04-12 · 7 min read #gemini-cli#google-gemini#terminal-ai#ai-agent#mcpAmazon Bedrock AgentCore Practical Guide: How to Build Secure Production Agents in 2026
A practical guide to Amazon Bedrock AgentCore for teams that need secure, production-ready agents, with clear coverage of Runtime, Memory, Gateway, observability, and rollout checks.
2026-04-12 · 6 min read #aws#bedrock#agentcore#ai-agent#mcpDeep Research Agents Practical Guide: How Developers and Knowledge Workers Should Use Them in 2026
A practical 2026 guide to OpenAI deep research for developers and knowledge workers, covering what it is, why it matters now, prompting patterns, MCP-connected workflows, pitfalls, and a simple decision checklist.
2026-04-12 · 7 min read #deep-research#ai-agent#mcp#chatgpt#research-workflowAI Gateway Platforms Comparison Guide: Vercel AI Gateway vs Cloudflare AI Gateway vs Amazon Bedrock AgentCore Gateway
A practical comparison of AI gateway layers as of 2026-04-12, showing where Vercel AI Gateway, Cloudflare AI Gateway, and Amazon Bedrock AgentCore Gateway belong in the stack.
2026-04-12 · 4 min read #ai-platform#ai-gateway#comparison#vercel#cloudflareBuilding Multi-Model Apps with Vercel AI SDK 6 and AI Gateway: A 2026 Practical Guide
A practical guide to building multi-model AI apps in 2026 with Vercel AI SDK 6 and AI Gateway, covering agents, MCP, provider routing, fallbacks, human approval, and a Next.js adoption checklist.
2026-04-12 · 7 min read #ai-platform#ai-sdk#ai-gateway#multi-model#fallbacksPydanticAI Practical Guide: Why Python Teams Adopt It for Production Agents in 2026
A practical guide to PydanticAI for teams that need Python-first agent systems, model portability, durable workflows, observability, and evaluation in production.
2026-04-12 · 6 min read #pydantic#pydantic-ai#python#ai-agent#mcp