I’ve been working as a software engineer for more than 15 years, and I’ve been working with coding agents for quite a while now. At first it was hard for me to accept that the rules of the code development game had changed forever.
Now, honestly, I’m pretty much surrendered to the quality of the code and reasoning these agents can produce. But there is still something I haven’t fully been able to feel.
I haven’t managed to feel that I’m working side by side with an engineer who knows the repository. Someone who is used to the project’s codebase, its strategies, its typical errors, the commands that should be run and the ones that shouldn’t.
I miss the feeling that the agent (I usually work with Codex and Claude, although mainly with Codex ) is a veteran teammate, not a rookie who has to review the whole repo, starting from the README and the Makefile, before writing a single line of code.
I thought it was all about refining prompts. Then I focused on operational memory, skills, MCPs, rules, global instructions, AGENTS.md, CLAUDE.md, and everything I kept reading over and over again in articles and posts.
I also had a “context” phase. I became obsessed with improving the context my agent was working with.
And yet I still had the same feeling.
The more I obsessed over prompts, memory, skills, and context, the more I started to feel that what the agent was missing was continuity. Something closer to what a teammate would ask on their first day at work:
Where were we? What did we do yesterday? What hypotheses did we discard?
Which file mattered? Which test was the right one?What should I not touch?
Where do I start?
It frustrated me to watch agents rediscover the repo, try overly broad commands, or attempt to run huge test suites that had nothing to do with the task at hand. So I started experimenting with operational continuity.
santism•1h ago
I’ve been working as a software engineer for more than 15 years, and I’ve been working with coding agents for quite a while now. At first it was hard for me to accept that the rules of the code development game had changed forever.
Now, honestly, I’m pretty much surrendered to the quality of the code and reasoning these agents can produce. But there is still something I haven’t fully been able to feel.
I haven’t managed to feel that I’m working side by side with an engineer who knows the repository. Someone who is used to the project’s codebase, its strategies, its typical errors, the commands that should be run and the ones that shouldn’t.
I miss the feeling that the agent (I usually work with Codex and Claude, although mainly with Codex ) is a veteran teammate, not a rookie who has to review the whole repo, starting from the README and the Makefile, before writing a single line of code.
I thought it was all about refining prompts. Then I focused on operational memory, skills, MCPs, rules, global instructions, AGENTS.md, CLAUDE.md, and everything I kept reading over and over again in articles and posts.
I also had a “context” phase. I became obsessed with improving the context my agent was working with.
And yet I still had the same feeling.
The more I obsessed over prompts, memory, skills, and context, the more I started to feel that what the agent was missing was continuity. Something closer to what a teammate would ask on their first day at work:
Where were we? What did we do yesterday? What hypotheses did we discard? Which file mattered? Which test was the right one?What should I not touch? Where do I start? It frustrated me to watch agents rediscover the repo, try overly broad commands, or attempt to run huge test suites that had nothing to do with the task at hand. So I started experimenting with operational continuity.