Tips for Interacting with GenAI Tools
— Life Upgrades — 2 min read
Embrace incremental improvements
The tools themselves won't necessarily change drastically overnight, but how well you use them can — finding a slightly better way to prompt or structure a task adds up to a real productivity gain over time, even without a new model release.
Use Chain-of-Thought prompting
Asking a model to "think step by step" before answering tends to noticeably improve both its accuracy and the quality of its reasoning, especially on anything more complex than a simple lookup.
Focus on system design and QA over typing speed
As AI takes over more of the routine coding work, the skills that stay valuable are the ones AI is worse at — designing systems well and catching what's wrong with them. Coding speed matters less than it used to.
Ask for options, not definitive answers
You'll usually get more useful results by asking a tool to lay out a few alternatives for you to weigh, rather than requesting a single "correct" answer up front.
Make space to experiment
Advocate for internal hack days at work, and keep side projects going in your own time — low-stakes environments like these are where you actually build fluency with new tools, without the pressure of it needing to work perfectly the first time.
Combine your existing expertise with AI
If you already have domain knowledge or programming experience, pairing it with AI tools tends to outperform either the tools alone or someone without that background — your judgment is what directs the tool well.
Keep the ethical questions in view
It's worth staying aware of where training data comes from and how AI adoption is reshaping traditional roles in your industry — these aren't settled questions, and it's reasonable to hold some discomfort alongside the productivity gains.
Source: Watch the original video
