Popular：
Virtualization DNS security formal verification reachability analysis compiler errors macro conflict web extension development framework Bitmap Graphics API inconsistencies All Tags

Programming with Agents: Beyond LLM Code Generation

2025-06-11

This article explores a revolutionary approach to programming using agents. The author defines an agent as a for loop containing an LLM call, granting the LLM access to compilers, the file system, and test suites. This contrasts sharply with programming solely with LLMs (akin to coding on a whiteboard), where agents, through environmental feedback, drastically improve code generation efficiency and accuracy. The author shares case studies of using agents for GitHub App authentication and handling JSON in SQL, demonstrating their power in boosting productivity and tackling complex tasks. While agents require more time and computational resources, their efficiency gains and potential for reducing human error position them as powerful tools for the future of programming.

Read more
(crawshaw.io)
Development agents

Programming with LLMs in 2024: My Experiences

2025-01-07

This post summarizes the author's experiences using generative models for programming over the past year. He found LLMs to be a net positive on his productivity, particularly for autocomplete, search, and chat-driven programming. While chat-driven programming requires adjusting workflows, it provides a first draft and facilitates quicker error correction. The author emphasizes that LLMs excel with well-defined problems and advocates for smaller, more independent code packages for better LLM interaction. He introduces sketch.dev, a Go IDE designed for LLMs to streamline the feedback loop and boost efficiency.

Read more
(crawshaw.io)
Development