Eric J Ma's Website

In this blog post, I share how coding agents like Cursor, GitHub Copilot, and Claude Code can access files across your entire file system—not just your current workspace. By simply referencing explicit file paths, you can pull code, documentation, or configs from any repository without complex workspace setups or copying files. I explain practical workflows and tips for cross-repository access, making coding and writing more seamless. Curious how this can simplify your development process and boost productivity?

In this blog post, I share how I discovered PocketFlow, a minimalist framework for building LLM-powered programs using graph-based flows instead of complex loops. By rethinking my approach, I replaced 307 lines of agent orchestration code with just 4 lines, making my agents more modular, clear, and easy to visualize. I walk through practical examples, show how to build and visualize agent architectures, and reflect on the benefits of graph-based thinking for LLM applications. Curious how this shift can simplify your own AI projects?

In this blog post, I share practical strategies for letting coding agents work autonomously while minimizing risks, like setting intelligent boundaries for command approvals, using plan mode, and writing prescriptive prompts. I also discuss real-world lessons learned from agent mishaps and offer tips for managing multiple agents safely. Curious about how to empower your coding agents without losing control?

In this blog post, I share how combining AI coding assistants with Marimo notebooks can supercharge your Python development and data science workflows. I walk through handy features like the `--watch` flag for live updates, the `marimo check` command for code quality, and even advanced options like MCP and built-in AI editing. Curious how you can automate and speed up your notebook workflow while keeping your code clean?

In this blog post, I share my first impressions of Anthropic's skills repository, comparing its token-efficient, customizable approach to the more standardized MCP server model. I break down the strengths and trade-offs of each, from creative workflows to technical utilities, and raise open questions about distribution and cross-vendor support. Curious about which approach might fit your workflow best?

In this blog post, I share what I learned building LlamaBot: the real challenge in AI-assisted development is keeping AI agents up-to-date with evolving documentation. I explain how the Model Context Protocol (MCP) lets LLMs access dynamic, queryable knowledge bases, solving the obsolescence problem and enabling smarter, context-aware AI assistants. Curious how you can make your documentation instantly accessible to any AI agent?

In this blog post, I share my hands-on comparison of DSPy and LlamaBot for building structured LLM applications, using a real-world expense extraction example. I explore how each framework handles schema design, type safety, and prompt optimization, highlighting their strengths and trade-offs. Curious which approach might best fit your next LLM project?

In this blog post, I share hard-earned lessons from using AI coding agents on real projects. I discuss why effective agent use goes beyond good prompts, highlighting the importance of systematic workflows, external memory, and fast iteration. I cover practical patterns for planning, testing, refactoring, and documentation, plus tips for integrating agents into your development process. Curious how these strategies can help you get the most out of coding agents?

In this blog post, I share how I solved the challenge of using multiple GitHub accounts on the same computer by configuring separate SSH keys and updating SSH and Git settings. I walk through step-by-step instructions, troubleshooting tips, and ways to automate account switching for different repositories. If you've ever struggled with Git pushing to the wrong account or want a smoother workflow for personal and volunteer projects, this guide is for you. Curious how to make Git always use the right account without hassle?

In this blog post, I share how using AGENTS.md—a new open standard for AI coding agents—lets you teach your LLM assistant project-specific preferences that persist across sessions. I cover practical tips like enforcing markdown standards, specifying test styles, and introducing new tools, all by updating AGENTS.md. This approach turns your agent into a trainable teammate, not just a forgetful bot. Want to know how to make your coding agent smarter and more aligned with your workflow?