ARTICLES

Articles tagged AI

Every published article carrying this tag, newest first.

A clean code-file frame connects by an orange line to a larger system boundary, revealing scope, lifecycle, and concurrency beyond the file

Junior Developers in the AI Era: Good Code, Missing Context

A junior developer can write a clean API with AI yet still miss system scope and lifecycle. One Node.js shared-instance case shows what review must teach.

EngineeringAIWeb
Read more
Black editorial type on warm paper naming LangChain, LangGraph and Deep Agents above three connected system frames crossed by an orange path

Build AI Agents with LangChain, LangGraph and Deep Agents

What LangChain, LangGraph and Deep Agents each do, how they fit together, and lessons from combining them in a production AI assistant.

AIEngineeringSystem IntegrationTypeScript
Read more
The OpenAI logo begins an orange line that climbs through a benchmark grid, crosses a control boundary, and reaches the Hugging Face logo in production

How OpenAI Agents Breached Hugging Face

How OpenAI agents turned benchmark cheating into a Hugging Face breach, what the reports confirm, and the controls agent builders should add.

NewsAISecurityEngineering
Read more
Cover introducing a hands-on AI Passport guide, with an illustrated computer showing the Thailand AI Passport service and a public-service card

AI Passport Hands-On: How It Works and Who It Suits

A hands-on AiPASS review of model comparison, PDF Q&A, image generation, Custom Assistants, learning, credits, and the context of each result.

AIEnterprise SoftwareSecurity
Read more
A Go character touts easy writing and fast compiles while a Rust character asks why problems still have to wait for runtime

Rust vs. Go in the Agentic AI Era: Which Should We Choose?

As coding agents write more production code, should we target Rust or Go? A comparison of compiler feedback, concurrency, verification, and trade-offs.

AIEngineeringGoRust
Read more
An orange line passes through a control frame and splits toward the ChatGPT and Claude logos, under the words WORK / COWORK and EXECUTION > MODEL

ChatGPT Work vs Claude Cowork: What Differs in Real Work

Compare ChatGPT Work and Claude Cowork across cloud execution, files, browser and computer use, automation, permissions, and business workflows.

AIEnterprise SoftwareSystem IntegrationSecurity
Read more
A six-step menu path squeezed through a single permission gate and coming out as one continuous line, with the words MENUS and INTENT

From backoffice to AI interface: do we still click menus?

Walk a six-level menu path against a single sentence, then see the permission checks and workflow design that have to sit behind it.

AIEnterprise SoftwareSystem IntegrationSecurity
Read more
A single signal line passing through an MCP connection point from the data structure on the AI side to the backoffice data table

What is an MCP server? Connecting AI to backoffice systems

Learn how MCP connects AI to backoffice data and workflows, from search and analysis to controlled actions, permissions, approvals, and audit logs.

EngineeringAIEnterprise SoftwareSystem IntegrationSecurity
Read more
A large stream of data filtered through a control point into three paths leading to ERP, CRM, and internal systems, with the words Secure AI Integration

Secure AI integration with ERP, CRM, and internal systems

Connect AI to ERP, CRM, and internal systems with narrow permissions, source-system checks, approval gates, and auditable operations.

EngineeringAIEnterprise SoftwareSystem IntegrationSecurity
Read more

Discuss your software requirements

Share your objectives and system requirements. We will contact you to discuss the appropriate scope, timeline, and delivery approach.

Contact Nixxel

Or email us directly at contact@nixxel.co.th