Designing Subagents in the Cloud: Agno vs LangChain vs AI SDK
A practical, source-audited guide to designing subagents for real cloud workloads, with concrete patterns and tradeoffs across Agno, LangChain, and Vercel AI SDK.
14 min read
Archive
Articles, page 4
Older writing on AI applications, software engineering, and developer tools.
From Chat to Agent UI: ChatKit, A2UI, and Structured Interaction Surfaces
Agent UI is becoming its own stack. Here's how ChatKit, A2UI, and MCP Apps fit together, where plain chat breaks down, and how to design structured interaction surfaces that actually help users get work done.
24 min read
How to Improve Code Quality During Vibe Coding
Vibe coding is fast, but speed without discipline creates debt. Here's how to keep your AI-generated code clean, tested, and maintainable.
13 min read
How to Stream LangChain and LangGraph into AI SDK
A source-audited, practical guide to building streaming APIs with LangChain and LangGraph, then consuming them cleanly with AI SDK from simple chat to durable agents.
19 min read
LLM-as-a-Judge for Agent Apps: Biases, Blind Spots, and Fixes
LLM-as-a-judge can be one of the most useful patterns in agent evaluation, but only if you understand where it breaks: order bias, self-preference, verbosity bias, weak judges, and evidence-free scoring. This guide explains the pattern, the common traps, and the fixes that make it practical.
17 min read
Trace Grading vs Scenario Testing: How to Evaluate Agents in Production
Why production agent evaluation is moving beyond output-only checks, how trace-aware grading complements scenario testing, and how LangWatch, LangSmith, and Langfuse compare.
12 min read
The Evolution of AI Agent Orchestration: System Prompts, Skills, MCP, and Plugins
How AI-assisted engineering workflows mature from a simple system prompt into skills, MCP tools, and full plugins — with real engineering examples at every stage.
19 min read
Computer-Use Agents for QA, Legacy Systems, and Browser Automation
A source-audited guide to where computer-use agents are already practical, where they still break, and how to deploy them safely for QA, legacy enterprise workflows, and browser automation.
12 min read