← sources

Recently, my company started trialling Claude Code. I don't know who effective it will be because higher-ups do not understand GenAI...

source: reddit (reddit:ClaudeAI)published 2026-10-20 14:33ingested 2026-10-21 02:36

retained text

I work on a legacy product which is developed/ tested in a proprietary environment.

My higher-ups have been working on this product for more than a decade. They believe they have job security until the product is completely phased out. As a result, they refuse to learn anything new.

We've been slowly integrating GenAI into our workflows. Dealing with these higher-ups is turning into a nightmare.

For a repetitive tasks, I shared a simple agentic workflow in a GitHub\* repo. One of them was like: "What the hell is GitHub? Where is the download button? This is too complicated! Just give an exe file."

(\*Everyone in the team has attended 3 session/ 9 hour workshop on GitHub and GitHub Copilot.)

The metrics they collect for GenAI doesn't make a lot of sense:

Top 3 models used

Tokens used (they just discovered tokenmaxxing)

AI contribution to a task in %

Time saved

Recently, my company started a pilot program for Claude Code, and we attended a couple of sessions on it. As of now, we have limited number of licenses that is being rotated. For example, 100 people attend an intro session and they get Claude Code license for a week. And then the license is moved to next 100 users. The higher ups are already complaining about how there is no GUI, everything has to be typed out, and Claude Code looks like some tool that was created in the 80s.

  
Not sure how to deal with these people...

document fetched 2026-10-21 00:21 · canonical https://reddit.com/r/ClaudeAI/comments/1xatrmg/recently_my_company_started_trialling_claude_code