| A founder I work with recently ran a small experiment with an engineering team at a Series B company that I thought was interesting; I'm not technical but was curious as to how this session would play out. The setup was simple: six engineers spent a few 2 days experimenting with AI in their real development workflows. Nothing theoretical — just trying things directly against their codebase and normal tasks. A few observations stood out. 1. Initial skepticism was high Several engineers were openly skeptical at the start. The general sentiment was something like: “AI coding tools are interesting, but mostly useful for small snippets.” "I wouldn't put my name on code that wasn't 100% mine" was another... one even said "AI is the devil!" Most of the team had tried assistants before but hadn’t really changed how they worked. 2. The shift didn’t come from code generation What surprised everyone was that the biggest changes weren’t about generating code faster. The moments where things “clicked” tended to be things like: exploring unfamiliar parts of the codebase debugging tricky issues quickly testing alternative approaches generating scaffolding to explore architecture ideas Once engineers started using AI more as a thinking partner rather than a code generator, workflows changed pretty quickly. 3. Behavior changed very fast once a couple of the engineers started to use libraries they hadn't used due to time taken to understand how to integrate it. After that, the rest of the team started experimenting with similar approaches. Within a week the company apparently saw a huge spike in requests for "AI-tool" licenses internally. 4. The interesting part wasn’t the tool What seemed to matter more was that the engineers had time and space to experiment together. A lot of teams seem to have access to AI tools now, but they’re still using them in fairly shallow ways. When people start experimenting with their real workflows, usage patterns seem to change very quickly. Curious if others have seen something similar inside their teams. Specifically: Where have AI tools actually changed engineering workflows for you? What use cases ended up being more useful than expected? Did adoption happen gradually, or was there a “flip the switch” moment? Would be interested to hear how this is playing out in other engineering teams. |
The other thing that made a massive difference was investing in project context files. Most teams use AI tools with zero project-specific context — the AI knows nothing about their conventions, patterns, or architecture decisions. It is essentially a smart stranger every session.
When you give the AI a well-written .cursorrules or similar context file that encodes your team's actual patterns — naming conventions, preferred libraries, error handling approach, testing philosophy — the output quality jumps dramatically. Instead of generating generic React code, it generates code that looks like YOUR team wrote it.
I have been maintaining cursor rules across 16 frameworks and the pattern is consistent: teams that invest 30 minutes upfront writing good context files get maybe 3-5x more useful output from AI tools than teams using them out of the box. That initial setup cost is what makes the difference between "neat toy" and "actually changed my workflow."
The social contagion effect you describe (one engineer starts, others follow quickly) is real too. In my experience it usually starts with someone sharing a particularly impressive AI-assisted debug session or refactor, and then everyone wants to know how they set it up.