site banner

Tinker Tuesday for May 13, 2025

This thread is for anyone working on personal projects to share their progress, and hold themselves somewhat accountable to a group of peers.

Post your project, your progress from last week, and what you hope to accomplish this week.

If you want to be pinged with a reminder asking about your project, let me know, and I'll harass you each week until you cancel the service

1
Jump in the discussion.

No email address required.

This is going to be, quite-possibly, a below-illiterate tech question, so please bear with me and save all openly expressed disdain for the end.

I run a small business that makes about $20K a year, as a side hustle. I started soon after AI hit the mainstream and have found the $20 a month tier of ChatGPT to be invaluable for streamlining administrivia type tasks such as boilerplate emails, plus helpful for very early brainstorming and having a minimally effective sales pitch in 15 minutes. I still do some amount of cleaning up for these processes, less for the boilerplate things, more for the creative things. I have trained a couple of GPT’s to be focused on the specific tasks I need them for, and will continue to do so as/if I expand.

My question is, am I missing out on some capability by only using the basic bitch version of GPT? Could I be getting more bang for my buck, better sales emails, better crafted first pass sales pitches, more automation, etc, by changing products? Should I use a different LLM company, or pay for API access, or buy a good GPU and train my own sandslave, or what? Or am I fine where I’m at?

Knowing what little we do about your side hustle, I think you're fine.

Paying for API access can be valuable if you have a lot of data you want to curate/generate. Imagine you are selling 10k items and want to take the specs + reviews to make better descriptions or something.

That's it though, you won't get much better quality from another model or benefit out of building your own rig. Yet.