Before LLMs became big, I used emojis in my PRs and merge requests for fun and to break up the monotony a bit. Now I avoid them, lest I be accused of being a bot.
Thanks @todsacerdoti for sharing sysp, really appreciate you noticing it! I'm a highly motivated engineer, with plenty of skills ranging from web development to embedded engineering, and would love to discuss potentially being a part of your team at Pipedream. I'm open to chat about roles/fit on X, @karans4_SF or you can contact me though my email on my website (check my profile).
Rather than using symbols, have you considered using common, self-descriptive words that would likely be one token? If the model hasn't been trained on the language, and the language itself isn't self-descriptive, then the language spec would have to be part of the context window too.
I can't stand the style as much as the excessive use of hyphens. The "It's not just ..., it's ...." every 5 sentences is too much once you notice it. However, every LLM seems to converge on this style. It wouldn't wouldn't write like that if it didn't work to some degree, so maybe it knows something we don't.
