I’ve been using ChatGPT as a sparring partner for a few weeks now to help ensure that my development work in analog electronic circuitry is successful. This is rather doubtful, because ChatGPT can’t consistently keep track of the information it’s been given. It often fails to honor the agreement not to make assumptions, so I have to keep correcting ChatGPT. But the biggest mistake has just happened. In a previous chat, a function was defined that was supposed to be carried over into a new chat. The function was described in a mathematical equation. In the newly opened chat, the equation was presented inside a symbol; ChatGPT analyzed the symbol and reproduced the equation incorrectly. Pointing out the error did not lead to a correction, however; the statements became increasingly convoluted. You can’t work with an AI like this. I’m going to cancel my subscription.

Welcome to the forum!

How to preserve memory across sessions and threads?

My best advice, based on what I have been using for more than a year, is to consider OpenAI Codex.

Although Codex is primarily known as a software-development tool, it is capable of much more.

Codex can now operate your computer alongside you, work with more of the tools and apps you use everyday, generate images, remember your preferences, learn from previous actions, and take on ongoing and repeatable work. The Codex app also now includes deeper support for developer workflows, like reviewing PRs, viewing multiple files & terminals, connecting to remote devboxes via SSH, and an in-app browser to make it faster to iterate on frontend designs, apps, and games.

When combined with MCP servers—including custom ones—skills, and slash commands, Codex can support workflows that would have been difficult to imagine with generative AI only a few years ago.

Note: Much of what you are seeking is covered in the linked pages. Each link leads to different information that may be useful for evaluating the available options.

Update

Record & Replay lets you demonstrate a workflow on your Mac and turn it into a reusable skill. Use it when the workflow is repetitive, depends on your preferences, or is easier to show than to describe in a prompt.

Note: I do not have a Mac running macOS, so I am currently unable to use this feature.