Blog

AI & LLMs

Articles on AI agents, LLM engineering, and applied machine learning — prompt design, RAG, evaluation, and building production AI features.

130 articles

AIOctober 202612 min read

A2A vs MCP: Which AI Agent Protocol Do You Actually Need?

MCP connects an agent to tools; A2A connects agents to each other. Discovery, state, long-running tasks, auth and governance compared in one table.

AIOctober 202611 min read

AI Agent Cost Optimization: Budget Per Task, Not Per Call

Agents loop, spawn subagents and grow their context, so per-call pricing math underestimates them. Measure cost per completed task, then control it with max_turns, subagent caps, tool search, compaction and model routing.

AIOctober 202612 min read

AI Agent Evaluation: How to Test Tool Calls and Trajectories

An agent can return the right answer after calling the wrong tool, with the wrong arguments, in the wrong order. Here is how to grade the trajectory itself, from production traces to promptfoo assertions against a NestJS-hosted agent.

AIOctober 202611 min read

ChatGPT Plugins (Apps SDK) Guide: Build an ERP Plugin on MCP

Connectors became apps, apps became plugins, and the Apps SDK URL now redirects to the plugins docs. Here is what a 2026 ChatGPT plugin actually is, and how to build one for an ERP order-status lookup.

AIOctober 202612 min read

Claude Managed Agents vs OpenAI Agents API: Full Comparison

Claude Managed Agents and the OpenAI Agents API both run the agent loop for you. A side-by-side of objects, sandboxes, custom tools, cron scheduling, ZDR, residency and pricing for teams retiring their own loop.

AIOctober 202613 min read

Google ADK 2.0 Tutorial: Graph Workflows and Multi-Agents

A hands-on Google ADK 2.0 tutorial: build a procurement approval graph with routes and human approval, persist sessions across restarts, choose between graph, dynamic and collaborative workflows, test it, and migrate from 1.x.

AIOctober 202613 min read

LangChain create_agent Middleware Guide: PII, Limits, Fallback

LangChain 1.0 replaced create_react_agent with create_agent and a middleware list. This guide migrates an old agent, then adds PII redaction, tool-call budgets, a closed-period guard, retries with model fallback and human approval, and explains why their order matters.

AIOctober 202612 min read

LangGraph vs CrewAI in 2026: Which for Production Agents?

The same three-step invoice workflow built in LangGraph and in CrewAI Flows, compared on control model, checkpointing, resume semantics, human approval, observability and learning curve.

AIOctober 202613 min read

Mastra TypeScript Agent Framework Tutorial: Workflows and Evals

A hands-on Mastra 1.x tutorial for TypeScript teams: a Zod tool scoped by RequestContext, an agent with structured output, a purchase-order workflow that suspends for a manager, quick-check evals in Vitest, Studio, and how it compares with the Vercel AI SDK and the OpenAI Agents SDK.

AIOctober 202613 min read

Microsoft Agent Framework 1.0: Semantic Kernel Migration Guide

Microsoft Agent Framework 1.0 replaces Semantic Kernel and AutoGen for new agent work. Here is how the concepts map, what the C# and Python code looks like before and after, and what the BUILD 2026 harness and the pre-release CodeAct package change.

AIOctober 202611 min read

OpenAI Agent Builder Shutdown: Migrate Before Nov 30, 2026

Agent Builder goes away on 30 November 2026 and Evals turns read-only a month earlier. An inventory, export and parity-test plan for moving each workflow to the Agents SDK or a Workspace Agent.

AIOctober 202612 min read

OpenAI Agents API Tutorial: Hosted Agents and Sandboxes

A hands-on Agents API walkthrough: create a session in an OpenAI-hosted sandbox, stream turns, steer, handle webhooks, clean up, and check residency first.

AIOctober 202611 min read

OpenAI Agents API vs Agents SDK vs Responses API: Which?

OpenAI now documents four ways to build an agent. A decision guide to the Agents API, Agents SDK, Responses API and ChatKit: who owns the loop, where state lives, ZDR, data residency, cost and lock-in.

AIOctober 202611 min read

OpenAI Agents SDK Guardrails: Input, Output and Tool Checks

The OpenAI Agents SDK has three guardrail layers that run at different points of a workflow. Here is when each fires, why parallel input checks can let a tool run anyway, and how a tool guardrail stops a journal entry posting to the wrong branch.

AIOctober 202611 min read

OpenAI Agents SDK Handoffs vs Agents as Tools Explained

A handoff gives the conversation to a specialist; agent.as_tool() borrows the specialist and keeps the conversation. That choice decides history, guardrails and tracing.

AIOctober 202612 min read

OpenAI Agents SDK Sandbox Agents: Manifest, Docker, E2B

SandboxAgent, the Manifest, capabilities, lazy skills, Unix-local versus Docker and E2B, snapshots and resume: give an agent a shell, not your laptop.

AIOctober 202612 min read

OpenAI Agents SDK Sessions: SQLite, Redis and Postgres Memory

Choosing a memory backend for the OpenAI Agents SDK: SQLiteSession, RedisSession, SQLAlchemySession on Postgres or server-side Conversations, plus the compaction and encryption wrappers, compared on multi-worker safety, retention and branching.

AIOctober 202612 min read

OpenAI Agents SDK TypeScript Tutorial: Build a Tool Agent

A TypeScript-first build with @openai/agents: zod-typed tools, a hosted file search, a handoff, streaming from a Next.js route, a maxTurns cap and promptfoo trajectory tests.

AIOctober 202611 min read

OpenAI Agents SDK vs Google ADK vs Claude Agent SDK (2026)

A buying guide to the three vendor agent SDKs in October 2026: languages, handoffs vs graphs vs subagents, sandboxes, sessions, evals and hosted runtimes.

AIOctober 202612 min read

OpenAI ChatKit Tutorial: Self-Hosted Agent Chat in React

Build an embedded agent chat with @openai/chatkit-react and the ChatKit Python server: the respond loop, the Store contract, widgets, attachments, auth and domain keys.

AIOctober 202611 min read

OpenAI DevDay 2026 for Developers: Sol, Ultrafast, Agents API

What OpenAI DevDay 2026 actually changed for developers: GPT-6.1 Sol at one-fifth of Astra's price, a 6x-priced Ultrafast tier, computer use in the Agents API, Bedrock Managed Agents, the Decisions API preview and Codex Cloud, checked against the docs.

AIOctober 202612 min read

OpenAI Dots Agents Explained: Always-On ChatGPT Agents

OpenAI Dots are always-on ChatGPT agents on GPT-6 Astra with their own cloud computer. What OpenAI confirmed, what the press got wrong, the Custom Rules and Auto-review controls, and how to pilot one in an ERP back-office.

AIOctober 202611 min read

OpenAI Evals Shutdown: Migrate Your Evals to Promptfoo

OpenAI Evals goes read-only on 31 October 2026 and shuts down 30 November. Export definitions, runs and rows now, then rebuild them in promptfoo and CI.

AIOctober 202613 min read

OpenAI Realtime Voice Agent over SIP with gpt-realtime-2.1

Build an OpenAI realtime voice agent on gpt-realtime-2.1: Agents SDK tools, handoffs, guardrails, a SIP phone line, the GA migration and call costs.

More in this category

Back to all posts