AI Agents
Autonomous agents, LLM applications, and intelligent systems
8 min readGPT-6 Astra AGI: How to Actually Use It and Get Its Best Performance
GPT-6 Astra is OpenAI's most capable model, but 'AGI' is a marketing claim. This guide shows you how to access, use, and get the best from it with practical endpoints, pricing, and agent patterns.
5 min readWhat is Codex starts encrypting sub-agent prompts? A Practical Overview
OpenAI Codex now encrypts sub-agent prompts by default. This change affects how agentic systems share context between sub-agents. Here's what it means for your AI pipelines.
4 min readBuilding reliable agentic AI systems: A Practical Overview
A practical guide to building reliable agentic AI systems covering structured outputs, observability, fallbacks, and cost controls with real code examples.
5 min readWhat is Harness engineering: Leveraging Codex in an agent-first world? A Practical Overview
Harness engineering is the practice of building structured, safe environments for AI agents to execute code. This post explains how to leverage OpenAI Codex in an agent-first world, with concrete examples, cost breakdowns, and honest tradeoffs.
4 min readWhat Is Failing Grades Soar with AI Usage, Dwindling Math Skills in Berkeley CS Classes? A Practical Overview
UC Berkeley CS professors report a sharp increase in failing grades alongside higher AI usage and declining math skills. This post breaks down the data, the likely causes, and what it means for hiring and education.
4 min readWhat is My Agent Skill for Test-Driven Development? A Practical Overview
An AI agent skill for test-driven development automates the red-green-refactor cycle. Here's how it works, what it costs, and when to use it.
1 min readWhat is Is AI ruining our skills? Early results are in – and they're not good? A Practical Overview
4 min readWhat Is AI's Multiplying Effect on Existing Technical Skills? A Practical Overview
AI tools multiply existing technical skills by automating boilerplate, accelerating debugging, and enabling faster iteration. This post breaks down the mechanics, costs, and best practices.
2 min readSkills vs MCP: When to Use Which in 2026
Skills teach the agent your workflow; MCP connects live systems. Here is when to use each — and the hybrid pattern HN devs settled on in 2026.
9 min readAI Agents in Production: Best Practices for State Management and Reliability
Building reliable AI agents in production requires robust state tracking, error handling, and recovery mechanisms. Learn the core architectural patterns.
11 min readDesigning Multi-Agent System Architectures for Enterprise Workflows
How to orchestrate multiple specialized AI agents to solve complex, multi-step business problems. Explore communication protocols and handovers.
8 min readTesting AI Agents: Designing Robust Mocks and Integration Test Suites
AI agent behavior can be non-deterministic, making traditional testing hard. Discover strategy mocks and assertions to test agent logic reliably.
7 min readHuman-in-the-Loop AI Agents: Designing Seamless Handover Interfaces
Agents occasionally fail or hit authorization barriers. Here is how to design UIs for human review, escalations, and interactive decision overrides.
10 min readMemory Architectures for Conversational AI Agents: Short-Term vs Long-Term States
A comparison of conversation buffers, summarized context, and vector database embeddings for persisting conversational memory in complex workflows.
8 min readTool Use in LLMs: Design Patterns for Reliable Agent Actions
Tool use is how LLMs take actions in the world. These design patterns make the difference between an agent that works reliably and one that hallucinates parameters and loops forever.
6 min readDevin vs Claude Code vs Copilot Workspace: AI Software Engineers Compared
Three tools claim to be AI software engineers. Here is an honest comparison of what each actually does well, what the benchmark numbers mean, and when to reach for each one.
6 min readAutoGen: Microsoft's Multi-Agent Framework Explained
AutoGen lets you build systems where multiple AI agents collaborate, execute code, and involve humans in the loop. Here is how it works, when to use it, and the real tradeoffs you'll face in production.
8 min readHow to Evaluate AI Agents: Beyond Task Completion Rate
Task completion rate alone misses most of what matters in agent evaluation. Here is how to measure trajectory quality, cost efficiency, error recovery, and build your own eval suite.
5 min readLangChain Complete Guide 2026: When to Use It and When Not To
LangChain is a powerful LLM framework, but its complexity is frequently criticized. Here is when it genuinely helps and when you should skip it entirely.
8 min readRunning AI Agents in Production: What Actually Breaks
Deploying agents to production reveals failure modes that benchmarks never show. Here is what actually breaks and the patterns that keep agents stable under real conditions.
8 min readLlamaIndex for RAG: A Practical Implementation Guide
LlamaIndex is purpose-built for RAG and document Q&A. Here is how its core components work and when to choose it over LangChain.
7 min readComputer Use AI Agents: What They Can Do in 2026
Computer use agents can click, type, and navigate a real desktop. Here is what the technology can actually do, where it still fails, and when it beats a proper API integration.
7 min readBrowser Agents: Automating Web Tasks With AI
Browser agents let LLMs control a real web browser to navigate, click, fill forms, and extract data. Here is how they work, when they are worth the cost, and when they are not.
7 min readMemory in AI Agents: Short-Term, Long-Term, and Episodic
Agents without memory repeat themselves, forget context, and fail on multi-session tasks. Here is how short-term, long-term, and episodic memory work and how to implement each.