Can't wait for 'Tell HN: how I make mid six figures as a prompt engineer'.
Absolutely. See also: https://promptbase.com And we're still in the early days.
Unwillingly considering whether the easy bucks are worth the greasy feeling.
111–120 of 153 posts
I tried a number of these generators a week ago (or so), all with the same prompt: "A child looking longingly at a lollipop on the top shelf" with pretty abysmal (and sometimes horrifying) results. I'm not sure if my expectations are too high, but maybe I was doing it wrong?
>the ball is positioned in such a way that the llama has no real hope of making the shot I love that we're at the level where the physical "realism" of correctly representing quadrupedals playing basketball is a thing now. I suppose the next level AI will be expected to model a full 3d environment with physical assumptions based on the prompt and then run the simulation
Can't wait for 'Tell HN: how I make mid six figures as a prompt engineer'.
"I started out as a patty inversion engineer at McDonalds."
This is kind of funny. DALL·E is one of the most impressive pieces of software, but such a basic feature like history is curiously underpowered.
Can't wait for 'Tell HN: how I make mid six figures as a prompt engineer'.
"We let our graphic designer go so we could onboard a AI Prompt Engineer" "How much are we paying him?" "About $225k plus bonus and equity" "And how much was the graphic designer paid? "$55k" "..."
How could all this play into "flooding" the NFT markets?
Film still of a llama in a jersey dunking a basketball like Michael Jordan, low angle, show from below, tilted frame, 35°, Dutch angle, extreme long shot, high detail, indoors, dramatic backlighting.
https://cdn.discordapp.com/attachments/999377404113981462/10...
Llama in a jersey dunking a basketball like Michael Jordan, screenshots from the Miyazaki anime movie
https://cdn.discordapp.com/attachments/999377404113981462/10...
Earlier quoted context omitted.
I suspect that many of the "impressive" examples that we see from tools like this have been carefully selected by human curators. I'm sure it's not at the level of "monkeys + typewriters = Shakespeare [if you're sufficiently selective]", but the general idea is still applicable.
Most of DALL-E2 output is great out of the box, the selection process is just fine tuning the results to create something the human in front of the computer likes. DALL-E2 can't mindread, so the image produced might not match what the human had in mind. There is however one thing to be aware of, the titles posted on /r/dalle2/ and other places are often not the prompts that DALL-E2 got. Instead they are a fun descrip…
Here's a result for the prompt of "Woman with green skin, leaves instead of hair, wearing a simple dress, far shot, digital art, hyper-realistic, 8k, ultrahd," for example (all four images)
You will note that none of them are even basically fulfilling the prompt, as well as all four being, in my estimation, ugly and uninteresting. That's not unusual for prompts that involve some element of the fantastic -- though there are corners of less-realistic digital art that it does do well.